Stop With the “Money Printing” Madness

I never stop seeing the term “money printing” all over the place.  It has to be the most abused term in all of economics and finance.  The madness must end!  So let’s try to make this so simple that a 6 year old could understand it.

1)  Banks create most of the money in our system.  Loans create deposits and deposits are, by far, the most dominant form of money in the economy.  So, if you want to say someone “prints money” you would be most accurate saying that banks print money.

2)  The government is a user of bank money.  When the government taxes Paul they take Paul’s bank money and redistribute it to Peter when they spend.

3)  If the government runs a budget deficit (taxes less than it spends) then Paul buys a bond from the government and the government gives Paul’s bank deposit (which he used to buy the bond with) to Peter.  Paul gets a bond which the government created in much the same way that a private corporation creates a bond when they issue corporate debt.   If you want to say these entities “print” financial assets then fine.  Corporations print stocks and bonds every day and you don’t hear the world exploding with hyperinflation rants because of it….

4)  When the Fed performs quantitative easing they perform open market operations (just like they have for decades) which involve a clean asset swap where the bank essentially exchanges reserves for t-bonds.  The private sector loses a financial asset (the t-bond) and gains another (the reserves or deposits).   The result is no change in private sector net financial assets.  QE is a lot like changing your savings account into a checking account and then claiming you have more “money”.  No, the composition of your savings changed, but you don’t have more savings.

5)  Cash notes like the ones you have in your wallet are created by the US Treasury and are issued to the Federal Reserve upon demand by member banks.  This cash is literally “printed” by the Treasury, but serves primarily as a way for banks to service their customers.  In other words, if you have a bank account you can exchange your bank deposit for cash from the ATM or the bank teller. Cash is preceded by the dominant form of money, bank money.  But it doesn’t get printed off the presses and fired into the economy as some would have us believe.

See, there’s no “money printing” in any of this unless you want to distort the role of cash in the economy or refer to lending and security issuance as money printing.  Yes, QE alters the composition of private financial assets, but that’s about it.  No real “money printing” there either.   So, next time someone goes off on a “money printing” rant just point them in the direction of these 5 easy to understand steps.

* Confused?  See the following pieces:

1.  Understanding The Modern Monetary System

2.  Understanding Inside & Outside Money

3.  Understanding Moneyness

4.  Where Does Cash Come From?  

Cullen Roche

Cullen Roche

Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services. He is also the author of Pragmatic Capitalism: What Every Investor Needs to Understand About Money and Finance and Understanding the Modern Monetary System.

More Posts - Website

Follow Me:


  1. Cullen As usual, an excellent post. I am not clear on one point. You say “Paul gets a bond which the government created in much the same way that a private corporation creates a bond when they issue corporate debt”. However, isn’t it true that ultimately the government bond is replaced at maturity by money created “ex-nihilo”. In contrast, the corporate bond must be “paid off” with “inside money”.
    Corporations can’t create net financial assets. So it perhaps was not your point, but don’t these transactions differ in this respect? Thanks much for any thoughts.

  2. “However, isn’t it true that ultimately the government bond is replaced at maturity by money created “ex-nihilo”.”

    Say for example, the government raised the bond payoff money through taxes. Then bank deposits would disappear from the tax payer and be used to pay the bond holder. Or the government could sell another bond, in which cases much the same thing happens, except an NFA is added to the private sector. Perhaps the confusing part is that technically the inside money gets converted to outside money (deposit disappears and bank raises then transfers reserves to Tsy) and then back again (Tsy pays bond holder by crediting his bank w/ outside money, and the bank credits bond holder’s deposit with inside money).

    “Corporations can’t create net financial assets”

    What about a commercial bank deposit? Or a corporate bond? Or stock? Or do you mean “net of any liabilities?” Nothing does that in our system*. There’s a liability created for every asset created somewhere.

    *except for coins, which are created by Treasury as assets.

    • I think we’re jumping through all kinds of hoops to avoid the obvious — that Treasury bonds will eventually be replaced by the Fed’s powers of creating money. (Heck we’re seeing it with QE.)
      I thought that was one of the main thrusts of MR — to reassure us that a currency issuer can’t default. We are not going to ‘pay down’ the debt in the example you used of the taxpayer paying the bond holder. We can issue as many bonds as we wish (barring the inflation constraint) because the Fed can buy them.
      And your other example — adding new bonds to replace the maturing bonds — just punts the date of money creation down the road.

      • “We are not going to ‘pay down’ the debt in the example you used of the taxpayer paying the bond holder.”

        I don’t agree. Say we start from all clear balance sheets. Now Paul take a loan, buys a bond, the gov spends the proceeds on Paul, and then taxes Paul 100% to pay off the bond. We’re left with

        1. No more gov debt (bond paid off)
        2. Paul has a deposit equal to his loan from the bank

        If Paul pays off his loan, we’re back to the beginning: all the balance sheets are again clear.

        It’s be similar to step 5. here:

        except that Tsy now taxes Person x $100k, and uses it to pay the principal on the bond held by Person y, and then Person y paid off Person x’s mortgage (unfortunately there’s the complication of a house and two people in my Example 8: say they got married), but if those steps were performed, we’d be back to step 1 (with just the house as an asset, but otherwise all balance sheets clear).

        There’s nothing inherent in the system that forces the Fed the replace Tsy debt held by the public with reserves. That may in fact happen, but there’s nothing structural to force it. That was my point.

      • “And your other example — adding new bonds to replace the maturing bonds — just punts the date of money creation down the road.”

        Yes, w/o any taxation or revenue generation, THAT’S the case in which we’re not going to pay down the gov debt.

        • This is the weak point of MR — sometimes it gets slippy on what the outcome will be.
          If a Treasury is a corporate bond, then it must be paid off by taxpayers. But a Treasury can always be redeemed by the Fed, so let’s stop making that comparison.
          So, in theory, the Fed can pay off the bond using newly created money. In reality, we certainly know this will be the case.
          Your two examples are so far outside the realm of possiblity that it weakens the MR argument.
          1. If you issue a bond to Peter and pay Paul, do you really imagine we’re going to tax Paul to pay back Peter. You’re going to send Paul a Social Security check and then take it back?1
          2. You say that without taxation or revenue generation ‘that’s the case in which we’re not going to pay down the gov debt.’ AGain, obviously this is not going to happen.

          • My all Paul (no Peter) verbal example above was meant to have Paul fill in for all of the non-bank private sector, so of course it’s not realistic. It’s not so outlandish to imagine that we could be in a situation where the gov is actually generating more revenue from the non-bank private sector than it’s spending… and thus paying down the debt (late 90s for example, prior to any QE). Of course it wouldn’t happen all at once (taxing Paul 100%), but that was just meant to illustrate the mechanism.

            More broadly, I was trying to address CharlesD’s questions, and that there’s really not such a huge difference between money used to pay down corporate debt vs gov debt. Yes, it’s ALL created ex-nihilo ultimately. That’s how our system works… no way around that. But I think he was imagining that it HAD to be “freshly” created somehow… when in fact it could be done with recycled (through taxation or bond issuance) existing ex-nihilo inside money.

            As to your point 2.: I guess I should have left that as “I agree” because that’s what I was trying to do with you. ;)

            Again, as to the specifics of our situation right now, you may well be correct that the Fed will end up holding on to a lot of Tsy debt until maturity. I’m agnostic on that… just pointing out that there’s nothing inherent in the system that forces it…. and in fact we can look to the not so distant past for examples that prove that can happen.

  3. The weird thing is that actually the “wealth” increases as a result of QE, because the amount of debt increases. Again, any one with basic knowledge of accounting can understand this.

  4. Cullen, I was surprised to see this today on Sumner’s site, especially after his article some months back entitled “Keep banks out of macro.”

    Ha! he’s actually addressing the “loans create deposits” idea… but in a classic Sumner strategy he says such questions are useless, and causality is hard to tell (though it may well be more one way than the other). My guess (it’s hard to tell because he’s so non-committal): he’s actually come around to that way of thinking, but refuses to acknowledge that it has any significance! ;)

    He warns his readers about asking bankers for an explanation!

    Well, now he can at least dip his toe in the water with some of his more prominent co-MMists who’ve acknowledged this long ago.

  5. Thanks Tom and others for your comments. What I meant to say was that I think a Treasury bond COULD be paid with money created “ex-nihilo”, unlike a corporate bond. As you point out, when a T-bond matures, our current accounting pays it with tax money or additional borrowing.
    But,as said, it COULD be paid for with money created “ex-nihilo”, as his happening with “QE”.

    • I see… but bank deposits are created ex-nihilo (inside money), and that’s really the only money it can get paid with “directly” (which is kind of a strange concept, since ultimately it’s actually gets paid “indirectly” through the bank with outside money as I described above).

      What I’m saying is there’s *perhaps* even less of a straight line between the Fed creating money ex-nihilo money and paying down the principal of a maturing bond, regardless of where it’s held.

      Check the balance sheets part of this comment:

      It’s kind of a long rambling msg on my part… I fell asleep writing it and didn’t actually post it till the morning! Ha!

      But it *might* help to answer you question somewhat. Keep in mind I’m not trying to track the history of every dollar, but this equation is an OK approximation for this purpose:

      non-bank held money (bank deposits & cash) = bank loans + Fed & bank held Tsy debt – TGA balance

      So in other words, this gives you a rough breakdown on the “components” of private non-sector bank “money” available to be taxed or borrowed from (through Tsy auction) with which to pay off the principal of any maturing Tsy debt, including those held at the Fed until maturity. To make it simpler, just assume that the TGA balance (Tsy’s Fed balance of outside money) is zero. So that gives you a breakdown of private money in terms of corresponding pools of debt. However, I’m not claiming that this money is somehow “backed by” these categories of debt. Still, perhaps you could say something to the effect that the following fraction of private non-bank money used to pay (anything) resulted from (in some sense) the Fed’s Tsy debt purchases (i.e. Fed ex-nihilo money creation):

      (Fed held Tsy debt) / (bank loans + Fed & bank held Tsy debt)

      So, if this ration turns out to be 0.3, for example (It’ll always be between 0 and 1), can you say that 30% of dollars used to pay the principal on maturing Tsy bonds (on average) came form ex-nihilo Fed money? I’m not sure, but maybe!

      • Tom, I’m Irish so be kind. I get the “no new FAs” in the system. but what is the fed “swapping”? they buy UST but with what? so USG debt has grown from $9T in late 07 to $16.4T end of Q1. I get the Fed “had to” step in(how and where is subject to debate) but at some point the economy has to start generating greater tax receipts right? And at a pace that can service the debt at higher rates, right? It seems these discussions get caught up in the trees without looking at the forest.

      • Tom thanks that was very helpful in explaining where the money comes from to pay off the bonds. Which is very helpful in explaining the world as it is (reality). I was just saying that my understanding is that the Fed could, for example, hypothetically buy all of the T-bonds with “ex-nihilo” money and then “retire the debt” by writing all the bonds down to zero and putting a large debit in their capital account. This sounds bad but it would just be accounting and have no real world effect. And, as we know, the composition of private sector financial assets would change but not the amount. And the debt is gone! While I realize this is really a bit off topic, it is important to me in discussing with others because I believe it is one way of demonstrating that the “debt” is not a burden on the future – since it could be hypothetically be retired with no consequences for the economy. Understanding that this is true is important for discussing the “debt” with others so any disagreement welcome (I don’t want to be spreading false information). I’m not dogmatic about anything. I’m trying to understand. As we know, “dogmatic” thinking is part of the national problem. Thanks.

        • I see… not sure if any laws would need to be changed to just “retire” debt like that. But I see what you’re getting at. The trillion dollar coin idea is a little different but perhaps has the advantage of actually being currently legal. Another idea is that we could allow (again a law change) the Tsy to overdraft the TGA. Or if the Fed & Tsy coordinated and the Tsy bought Tsy debt directly from the Fed… essentially making the Fed a desk in Tsy. Then retiring the debt would not even be necessary, but easier if we wanted to do that.

          You’re describing the M M T idea of state money. Cullen has written about that saying maybe it’s an OK concept, but emphasizing it’s not what we have.

          • Thanks Tom Yes as I understand it, the law requires the debt to equal the cumulative deficit. It is a relic of gold standard days. I am not recommending to retire the debt. As Cullen has pointed out, it serves many useful putposes. It was just a way, with a hypothetical, to demonstrate that the “National Debt” is really just a private savings account and could hypothetically be converted to “cash” without changing the amount of financial assets in the system. And yes, this is a hypothetical not what actually is happening, which makes MR preferable to MMT in that sense.

  6. Regarding the phrase “money printing”: my take is it’s a way to shut down a conversation by trying to cause a negative gut reaction in those that hear it. This is done all the time, regardless of justification. Here are some other favorite phrases used like this: “central planning,” “central planners,” “collectivists,” “class warfare,” “legalized counterfeiting” (for fractional reserve banking), “helicopter money,” “government bureaucrats,” “government ,” “redistribution of wealth” (or just “redistribution”), “Keynesian” (for anything involving any kind of stimulus or deficit spending), “taking property by force” or “by gunpoint” or “legalized theft” (for taxation), “socialism” (used for anything involving gov spending), “government regulations,” “red tape,” “nanny state,” “statist,” “social engineering,” “pork barrel,” “waste fraud and abuse,” etc. Those are some favorites on the libertarian or right wing side of things anyway. The left wing has some favorites as well: “rentiers,” “rent taking,” “financial capitalism” (as opposed to “industrial capitalism” which is more noble), “the 1%,” vs “the 99%,” “banksters,” “neo-liberal” (believe it or not, that’s become a bad word in some quarters.. along with “privatization”), “plutocrats,” “oligarchs,” “too big to fail/jail” etc. “Corporate welfare” is used by both sides, and more frequently “crony capitalism.” These are more or less center left and center right gut-reaction phrases, but of course there’s even scarier ones on the extremes: “bourgeois,” “counter-revolutionary,” “class struggle,” “imperialist,” “oppressor,” “lackey,” “capitalist” (for anything even slightly right of radical left), “Zionist,” “bolshevik,” “Marxist” (for anything slightly left of the extreme right), “fascist” (for either slightly right or left of the favored extreme position). And then of course a few social ones too “Bible thumper,” “fundamentalist,” “reactionary,” “sharia law,” “wingnut,” “teabagger,” “illegal immigrant” (or just “illegal”), “secularist,” “welfare queen,” “food stamps,” “racist” (for even the slightest deviation from PC… both sides use it too), “community organizer,” “elitist,” “redneck,” “gun nut,” “media elite,” etc. Then there’s a couple of ones used w/in a group: “RINO” and “firebagger.” You ever hear of “firebagger?”… it’s for fans of “fire dog lake” which is considered a little too left purist for others on the left. That was a new one to me.

    Now some of those phrases can be perfectly legitimate and useful even, but when over used or used indiscriminately or obviously used to produce that gut level reaction, I always find it off putting…. except of course when I’m doing it… then it’s totally justified ;)

    • EXACTLY, And So SO AS IT with the DEMOCRATS in Power. Look how they are strong arming all they can. hell Money, Finance politics… It’s ALL THE SAME METRIC when Human are involved.. :)

      • I tried to parcel out blame in equal measures, you think I was too hard on the right?… or are you just demonstrating some of my logic in the final line? :)

        • You did fine, Funny, just before you typed that I was having a conversation with a family member about stock values returning to a mean after extremes and comparing that to other events of human endeavors.
          Seems like most things we do go from one extreme to the other or sort of a regression toward the mean.

  7. Sorry, no one on this page has succeeded in explaining this in a way that a six-year old could understand. And sadly, that’s why rants about money printing will continue. But please keep trying. It’s a worthy goal.

    • Cirrus, it’s because there needs to be a chart for visual and then about a 6 to 10 word statement regardless if it’s factual. Then people will retain it.
      Case in point, I just heard this morning on ABC news business about how the FED is Jucing the DOW and they put up a chart just like this article on FOX News:

      So people need a visual and a one line statement I think, to make it adult proof. :)

  8. “4) When the Fed performs quantitative easing they perform open market operations (just like they have for decades) which involve a clean asset swap where the bank essentially exchanges reserves for t-bonds. The private sector loses a financial asset (the t-bond) and gains another (the reserves or deposits). The result is no change in private sector net financial assets. QE is a lot like changing your savings account into a checking account and then claiming you have more “money”. No, the composition of your savings changed, but you don’t have more savings.”

    It can be money printing eventually because the depositor can then request his cash and the bank will ask for more money printed. So you are playing with words here and that is not a good sign. QE is not direct money printing but a lot of money may be printed because of QE that would not have been printed should QE had not taken place. In other words, QE is an indirect means of facilitating money printing by replacing toxic assets with clean deposits.

    Get it or you need more explanations?

    • Cash, however, is just a convenience for us private non-bank entities: First of all, in almost ALL circumstances it originated from a bank deposit (inside money… think ATM withdrawal). Sure you may have been paid by your friend in cash, by he got it from an ATM, or the person he did got it from one.. or from the cashier at the bank, etc. Trace it back far enough, it came from a bank deposit. And when that cash goes back to the bank… the same thing happens in reverse: it goes right back to being a bank deposit. So cash is just a temporarily suspended bank deposit… transformed into paper until it goes back to a commercial bank. That’s it! Nothing mystical about cash.

      The Fed will sell cash to banks when they need more of it for the convenience of their customers. What do banks pay for it with? With electronic reserves. While the cash sits in the banks’ vaults it’s just another form of reserves. Exactly equivalent. When the bank has too much cash on hand and wants to exchange it for electronic reserves, it can sell it back to the Fed anytime.

      We could eliminate physical paper cash and metal coins ANYTIME and not significantly affect QE. What you’re saying gives paper money some special significance. It doesn’t have any actually. It’s purely for convenience, and we could get rid of it tomorrow and not change anything, except our convenience level.

    • Here’s a simple example demonstrating that physical cash no special significance, say we start with a Treasury (Tsy), Fed, bank, Person x, all with absolutely clear balance sheets:

      Tsy, Fed, bank, x:
      Assets (A): $0
      Liabilities: $0

      Now say person x takes a loan from the bank for $100 and uses it to buy a T-bond, and then the Fed buys the T-bond from person x during QE.

      A: $100 Fed deposit
      L: $100 T-bond issued

      A: $100 T-bond & $100 loan of reserves to bank
      L: $100 reserves at bank

      A: $100 reserves (Fed deposit) & $100 loan to x
      L: $100 reserve loan from Fed & $100 deposit for x

      Person x:
      A: $100 deposit at Bank
      L: $100 loan from bank

      Person x could now withdraw his deposit in cash. But in order to do that the bank must first obtain cash from the Fed. The Fed already has a stack of cash in inventory which does not appear on its balance sheet because cash is essentially worthless when held by the Fed: it’s literally just paper when held at the Fed. It only takes on its facee value when sold to a bank, at which point it becomes a liability of the Fed and is entered onto the Fed’s balance sheet. This is probably the concept which confuses people: that paper money held at the Fed is worthless.

      A: $100 Fed deposit
      L: $100 T-bond issued

      A: $100 T-bond & $100 loan of reserves to bank
      L: $100 paper money distributed

      A: $100 vault cash & $100 loan to x
      L: $100 reserve loan from Fed & $100 deposit for x

      Person x:
      A: $100 deposit at Bank
      L: $100 loan from bank

      Notice what happened to what I called “reserves” before? Those were electronic reserves. They disappeared. Electronic reserves are Fed deposits (called “reserves” when in commercial bank Fed deposits). The bank paid for the paper money with its reserves, so the Fed erased the bank’s electronic Fed deposit, and sent it some cash (at which point the cash took on value). The Fed just replaced one liability for another. The bank just replaced one asset for another, both of which are classified as “reserves” (electronic Fed deposits and vault cash held at commercial banks are both “reserves”). Now when Person x withdraws his deposit as cash we have:

      A: $100 loan to x
      L: $100 reserve loan from Fed

      Person x:
      A: $100 cash
      L: $100 loan from bank

      Nobody else’s balance sheet changed. When this $100 is redeposited by x (a stand in here for all of the non-bank private sector) because it’s now more convenient to go back to an electronic bank deposit (again, assume is actually all the non-bank sector, so it could have changed hands several times), this process is exactly reversed, and then bank sells the $100 cash back to the Fed, and the Fed then retires this cash, and re-enters a $100 electron deposit for the bank as a liability in it’s place.

    • I forgot to add the $100 Fed deposit at Tsy to the liabilities of the Fed in the above sets of balance sheets! So the Fed has a total of $200 in assets and $200 in liabilities in both sets!

  9. Cullen, this is hard stuff for the layman to understand. I am one.

    Suppose I’m XYZ bank. I buy $10 billion of T bills which give me a small return which I keep in my reserves. The gov spends my $10 bn. Is it fair to say that the principal of the T bills is an inert financial asset whose value I cannot use to chase a return?

    Now, the Fed gives me $10 bn cash for my T bills. Where was that cash before the Fed gave it to me? It is my understanding that the Fed simply increases its balance sheet – in other words, they give me money that had not been previously loose in the economy being spent or chasing a return.

    So now, isn’t there the banks original $10 bn being spent by the gov, $10 bn in reserves that can now be used to buy more T bills or other qualifying assets (that couldn’t be purchased without QE), and the Fed holding a $10 bn claim against future revenues of the US gov?

    So if it’s not money printing, are we at least increasing the number of assets in the economy that chase return?

    • Here’s the starting balance sheets:

      Step #1: Initial balance sheets:

      Tsy, Fed, XYZ bank, non-banks:
      Assets (A): $0
      Liabilities (L): $0
      Equity (E): $0

      Step #2: Tsy auctions $10B in T-bills, XYZ buys them:

      A: $10B Fed deposit
      L: $10B T-bills auctioned

      A: $10B reserve loan to XYZ
      L: $10B deposit for Tsy

      XYZ Bank:
      A: $10B T-bills
      L: $10B reserve loan from Fed

      Step #3: Gov spends its Fed deposit on non-banks (assume they all bank at XYZ):

      A: $0
      L: $10B T-bills auctioned
      Negative Equity: $10B

      A: $10B reserve loan to XYZ
      L: $10B deposit for XYZ

      XYZ Bank:
      A: $10B T-bills + $10B reserves
      L: $10B reserve loan from Fed + $10B deposits

      non-banks (assume they all bank at XYZ):
      A: $10B deposit
      L: $0
      E: $10B

      Step #4: Now assume that XYZ bank uses the $10B in reserves on its balance sheet to repay the reserve loan it took to buy the T-bills (only the Fed and XYZ balance sheets change):

      A: $0
      L: $0

      XYZ Bank:
      A: $10B T-bills
      L: $10B deposits for non-banks

      Step #5: Now the Fed buys the T-bills (again, only the Fed and XYZ balance sheets change, but I’ll write them all down):

      A: $0
      L: $10B T-bills auctioned
      Negative Equity: $10B

      A: $10B T-bills
      L: $10B reserves at XYZ

      XYZ bank:
      A: $10B reserves
      L: $10B deposits for non-banks

      A: $10B deposits at XYZ
      L: $0
      Equity: $10B

      Notice that at Step #3, the non-banks get $10B in equity from gov spending. This is not changed by steps #4 or #5 (#4 was a step I inserted … the original Fed loan repayment,… just to keep the balance sheets tidy, but #5 was yours).

      In other words, all that happened was that deficit spending (defined as the Tsy auctioning bonds and then spending the proceeds… i.e. everything through step #3) put $10B in equity into the private sector. Everything after this step did NOT change anybody’s equity, including the private sector. That’s why deficit spending adds equity to the private sector, but QE does not! Here’s a case demonstrating QE wherein Tsy buys Tsy debt from non-banks instead of banks:

      Again notice that QE did not change anybody’s equity. Here’s an example of deficit spending with the non-banks buying the Tsy debt instead of the banks doing it:

      Notice that the non-bank’s equity increased by the exact same amount that Tsy’s equity decreased.

  10. “A 6 year old could understand it”?????

    Let this economic ignoramous give it a try:

    1. A bond is a fancy name for a loan (you have to pay the money back with interest).
    2. “Loans created deposits, and deposits are…the most dominant form of money…”
    3. This would mean the the U.S. is creating huge deposits with the selling of it’s bonds.
    4. How does the government intend on repaying these deposits/money?
    a. future income (taxes)
    b. the implied default of printing money

    Therefore, while the U.S. may not be printing money, it is guaranteeing it will do so if necessary.

    Cullen: What is the difference between printing money, and guaranteeing that you will do so as a last resort?

    • Bank loans create deposits… not ALL loans! If I loan you my car, or $10, that doesn’t create a deposit. In our system, only commercial banks and the Fed can create deposits ex-nihilo through the act of lending. The Fed creates outside money and the commercial banks inside money deposits. The Fed can go one further and create deposits through the act of buying stuff (as in QE). So Treasury doesn’t create net deposits when it sells bonds, nor do corporations. Treasury doesn’t have the ex-nihilo deposit creation ability.

      Check out a few of the examples here:

      Examples 8, 4, & 1.1 are a good place to start.

      • So if a government bond isn’t going to be issued in money, what’s it going to be issued in – promises?
        (Remember that gauranteed repayment?)
        And while you’re explaining this, remember that you’re speaking to a 6-year old.

        • OK, but first, just to clarify, say we start with Tsy, Fed, bank, Person x, Person y, and everybody’s balance sheet is clear:

          Fed, Tsy, bank, x, y balance sheets:
          Assets: $0
          Liabilities: $0
          Equity: $0

          Now say the non-bank takes a loan from the bank:
          Bank BS:
          Assets: loan to x for $100
          Liabilities: $100 deposit for x

          Person x:
          Assets: $100 deposit
          Liabilities: $100 borrowing from bank

          So at this point the bank has created $100 of deposits in inside money ex-nihilo for person x. Now lets say person x loans that $100 to person y by writing her a check, which she deposits:

          Assets: $100 loan to x
          Liabilities: $100 deposit for y

          Person x:
          Assets: $100 IOU from y
          Liabilities: $100 borrowing from bank

          Person y:
          Assets: $100 deposit
          Liabilities: $100 IOU to x

          So do you see the difference between the bank loan and Person x’s loan? No deposit was created with person x’s loan: it just changed hands from x to y. Other than the banks, only the Fed creates dollar denominated deposits like that. Sure an IOU can be in dollars, but it’s not equivalent to dollars. The banks’ and the Fed’s IOUs are essentially the same as dollars, and they are created in much the same way you’d write an IOU to a friend: from thin air. If you view all these balance sheet manipulations as exchanges of IOUs between the various parties… the deposits (Fed or bank) are special… they are actually completely equivalent to dollars.

          Now a Tsy bond issued by Tsy is NOT a dollar denominated deposit, but it is a safe asset (in that it won’t be defaulted on)… however its value can go up and down with changing interest rates. The holder must wait to collect the principal back, or risk selling in the near term for a loss. The interest compensates for this risk.

          Practically speaking, you can say a Tsy bond is close to money because it is so safe and can always be sold, but it’s not as much like money as deposits (which are money). So in our example, lets say person y buys a Tsy bond through Tsy direct:

          Assets: $100 Fed deposit
          Liabilities: $100 bond issued

          Assets: $100 reserve loan to bank
          Liabilities: $100 reserve deposit for bank

          Assets: $100 loan to x
          Liabilities: $100 reserve loan from Fed

          Person x:
          Assets: $100 IOU from y
          Liabilities: $100 loan from bank

          Person y:
          Assets: $100 Tsy bond
          Liabilities: $100 IOU to x

          The bank (bank for both x and y) had to be an intermediary because inside money does not go directly into the outside money deposit of the Tsy. The bank can create inside money (bank deposits) but it cannot create outside money (Fed deposits). What happens when y’s inside money goes to by the Tsy bond is that the bank borrows outside money from the Fed, which in turn creates it ex-nihilo, and then sends that to Tsy on y’s behalf to pay for the Tsy bond. In return, the bank is allowed to then erase y’s inside money deposit. So essentially, the inside money got transformed into outside money in this process. The opposite happens when Tsy spends the money.

          So those are all the relationships… all those items on the balance sheets represent “promises” of varying risk levels: either promises given or promises accepted, with the IOU from Person y probably being the most risky to accept. Does that help?

          The following example is similar and follows the process through a few more steps, including spending of the deposit by the Treasury and some of the loan re-payments:

          So again, the only entities in the above example creating dollars are the bank (inside money) and the Fed (outside money). Everyone else is stuck with creating something else: Tsy bond, IOU, etc.

          • In that 2nd set of balance sheets, the Fed balance sheet should have been:

            Assets: $100 reserve loan to bank
            Liabilities: $100 reserve deposit for Tsy

            The error was on the “Liabilities:” line. The correction was to replace “bank” with “Tsy.”

          • You’ve got to be shitting me. Did you run this explanation past your 6-year old son or daughter (or niece or nephew) before giving it to me?
            Cullen is wrong (yes CR worshippers – WRONG). This cannot be explained to a 6-year old.
            Bottom line: the U.S. can and does create money out of nothing. How you wish to obscure that fact – and to obscure the fact that all we have is a faith that the dollar will purchase something – is your business (and your self-deception), not mine.
            Sorry, I have a short fuse.

            • We’re not printing money. We’re printing net financial assets that have 99 percent moneyness and are being redeemed every month by electronically created 00s by the Fed.
              But we’re not printing money.

              • Electronically created 00s that have next to no long term inflationary impact on the economy. Details matter Johnny and troll…

              • Bonds aren’t 99% money. How could you assume that? Cullen is pretty clear that bonds and most financial securities are money-like, but not money in the sense that neoclassicals define something like the monetary base.

            • Sorry troll, I’ll try once more. You’re absolutely correct, it’s all promises: they are either written and exchanged for real goods and services or they are exchanged for other promises. It all starts off like this, where nobody owes anybody anything:

              John is owed nothing. John owes nothing.

              Sue is owed nothing. Sue owes nothing.

              This can be re-written as:

              John has no assets. John has no liabilities.

              Sue has no assets. Sue has no liabilities.

              Now let’s say John does some work for Sue, then she could pay him with an IOU:

              John has Sue’s IOU. John owes nothing.

              Sue has nothing. Sue gave John an IOU.

              That’s how someone get’s paid for their work with IOUs. They could also get paid for their real property (house, car, etc.) this way. Either way notice that if John and Sue were consolidated into one person (JohnSue), then Sue’s IOU would cease to exist because JohnSue would owe that to JohnSue.

              Alternatively, John and Sue could just exchange IOUs:

              John has Sue’s IOU. John gave Sue an IOU.

              Sue has John’s IOU. Sue gave John an IOU.

              That’s what it all boils down to! Double entry accounting: assets (what’s owed to you) on the left, and liabilities (what you owe) on the right. When an IOU goes back to its originator (e.g. when John returns Sue’s IOU to Sue) it’s destroyed (e.g. Sue rips it up). It all follows that same basic pattern. It ALL comes from nothing (all the financial assets, including money). Consolidating everyone’s records together, the IOUs always add to zero. Note that real assets (houses and cars) don’t go away, just the financial assets. That’s how our system works. Other than that basic idea, here are the other rules of the system:

              1. The Fed’s IOU is special. It’s a Fed dollar. ONLY the Fed can write this kind of IOU. Fed dollars ALWAYS represent what the Fed owes. Like all IOUs, when Fed dollars are returned to the Fed, they are destroyed.

              2. The banks’ IOUs are special. They are bank dollars. ONLY banks can write this kind of IOU. Bank dollars ALWAYS represent what a bank owes. When bank dollars are returned to the banking system, they are destroyed. All bank dollars are equivalent no matter what bank they came from, and they are completely transferable between banks (i.e. one bank will be willing to take over an IOU from another bank if the originating bank pays it in Fed dollars to do so).

              3. Fed dollars and bank dollars are both dollars and they are equal but they are not the same thing. Only the Fed can create and destroy Fed dollars, and only the banks can create and destroy bank dollars.

              4. Treasury can ONLY have Fed dollars (not bank dollars). It CANNOT get them by exchanging its IOUs (Treasury bonds) to the Fed for Fed dollars. That’s prohibited! It can NEVER spend more Fed dollars than it has: that’s also prohibited. Nor can it ever borrow Fed dollars from the Fed directly, but the Fed can obtain Tsy IOUs by writing IOUs to banks in exchange for Fed dollars they have. Likewise, the Fed can trade Tsy IOUs to banks in exchange for returned Fed IOUs (which the Fed then destroys).

              5. Everyone else: you, me, all individual people, and all non-bank businesses can have bank dollars, but we CANNOT have Fed dollars. We can get them from each other or from banks. That’s OUR money!

              6. Banks can get Fed dollars from each other, from Treasury, or from the Fed. Fed dollars are money for the banks and Treasury.

              7. Banks act as intermediaries between non-banks (people and businesses) and the Treasury. E.g., if the Treasury wants to pay you for work you’ve done for the government, it pays your bank in Fed dollars, and the bank in turn writes you an IOU of bank dollars. Likewise, when you pay your taxes, you end up returning your bank dollars to the banking system, and the banks pay Treasury your taxes in an equivalent amount of Fed dollars.

              8. Banks also act as intermediaries between non-banks (people and businesses) and the Fed. This happens in much the same way as is done in item 7. above with Tsy.

              I’m oversimplifying a bit, and I haven’t mentioned cash (which is a minor issue… just another IOU!), but that’s pretty much it! It’s all a system of IOUs, but not all IOUs are equal. Some IOUs are exactly money, and there’s two perfectly equivalent but separate forms of money: Fed money, representing what the Fed owes and which is traded between banks and between banks and Treasury, or traded back to the Fed (where it’s destroyed). And bank money, representing what a bank owes, and traded between people and businesses or traded back to the banking system (where it’s destroyed).

              The point is that “money” in our system is not like a house or a car or a pile of gold or any other permanent asset. It’s a promise, and it’s destroyed when returned to the entity which wrote it as an IOU.

            • I am only wrong if you think bonds are money. In that case, we might as well start calling all financial securities money. In which case there’s no need for any company or entity at all to issue financial assets because they can create their own money….Obviously, that’s wrong so let’s all get on the same page here and stop referring to bonds as money in the same sense that a bank deposit is money. It’s not and stating that bonds have the same level of moneyness as bank deposits is a very confused portrayal of the monetary system.

        • A bond is not “money” in the neoclassical sense as others have pointed out. A bond is a type of security that cannot be used for final means of payment. Something with the highest level of moneyness MUST be able to meet that one purpose. Things like stocks and bonds do not meet this criteria and must therefore be exchanged into something with a higher level of moneyness. Do not misconstrue bonds as money or you will incorrectly portray the primary role of what most people believe “money” is. Bonds are money-like, but are not perfect substitutes for something like a bank deposit. The US govt issues bonds because it must obtain money from the private sector. They do this for the same reason that a corporation issues a bond. The issuing entity wants the thing with higher moneyness so it issues a money-like instrument to entice others to give it something with higher moneyness. This is a basic fact of security issuance and so confusing bonds with bank deposits is a rather egregious error that misportrays the reason why something like bonds exist in the first place. If the US govt really just issued all its own money there would be no need for bonds in the first place. It doesn’t do that though. It issues bonds so it can obtain money from people who are willing to let the govt use their bank money.

          You could argue that QE “monetizes” the bonds when a non-bank sells it to the Fed, but again, you must be very careful to explain this process correctly to avoid the inevitable “money printing” and hyperinflation shrieking. You must also avoid the idea that the govt is the issuer of the deposit as it is most definitely not.

  11. It’s not ‘money printing’ but ‘net financial asset printing’, in the form of T-bonds, which have a scale of money-ness at something close to 100 percent.
    I’ve never understood the insistence in here that we’re not money printing.
    On one hand, MR has shown that we can print these net financial assets at the current rate without danger of inflation or default but then it won’t admit this is what we’re doing.
    If you say that a bond must be paid back by the taxpayer, you get hooted off the board, but in this debate the bonds are being described as similar to corporate bonds, in which they will be paid back by the taxpayers. Which is nonsense, because they are already being redeemed by the Fed in QE operations from electronically created money.
    I can’t quite figure out why the deception. What’s the long-term goal here? Promote the buildup of federal debt until the only way to finance spending is by directly issueing money?

      • Off topic perhaps but it has occurred to me that most “traditional” voices are worried about the Treasuries piling up on the asset side of the Fed’s balance sheet. Many of the same people are, of course, worried about the piling up of Treasury securities on the liability side of the Treasury’s balance sheet (the “National Debt”). For that portion of the Treasury debt which the Fed has purchased, haven’t both “worries” disappeared? That is, on a consolidated “government” basis, the Fed’s assets equal the Treasury’s liabilities and they would net to zero (for the portion of the debt owned
        by the Fed). Am I missing something? Thanks.

        • You are correct. However, I think that MR finds it less useful to consolidate the Treas and Fed balance sheets though… but that is a common assumption with M M T. The Fed is treated as an independent entity in MR, and there’s some good justification for that. However, it’s also clear that it’s not completely independent.

          • We separate the Fed and Tsy because that’s how the current legal structure deems it be. MMTers like to say that money is a creature of law and the laws in the USA very clearly created the Fed as an independent entity. A key aspect of the Fed’s independence is its role as a firewall between banking and govt as the creators of money. The Executive Branch in the USA (and Tsy by extension) DOES NOT have a monopoly over money. Consolidating the Fed into the Tsy mangles that relationship and is often used by MMTers to imply that banks are simply agents of the govt and essentially part of the govt. They act as though we have a nationalized money system when it’s anything but.

            For accuracy in descriptive understanding it is absolutely imperative that one not distort the powers of the executive branch as they are currently designed by law. MMT’s description is largely void of value due to this important misrepresentation of the institutional structures in place. All MR does is accurately describe the actual institutional structures.

  12. I “believe” that Cullen has explained this to me: (1) Uncle Sam’s Deficit Spending increases net financial assets, (2) the Fed is a downstream process marketing the Treasury Bonds/Bills and Mortgage Backed Securities that have already been created, and (3) thus the Fed does not “create” or “print money”. What MR says is that the Bonds held by the Fed and others will be “paid off” or “rolled over”, because (I believe), that has been Monetary Reality so far. My understanding is that MR does not predict things that might happen in the future. MR explains how things work today. So… (my point): What will the Fed do with all those bonds (bought under QE), down the road is unknown therefore beyond the scope of MR, and thus in vacuum of speculation — which we are free to engage in until our air has been sucked out.

    • I think you’ve got it! By “net financial asset” I always read that as “net of liabilities” i.e. “equity” and of course it means in the private sector that they increase. Due to the nature of double entry accounting, if ALL balance sheets (public and private) are consolidated you get a great big $0 all the time of course!!

      Your (3): yes, that is the sense in which there’s no “money printing.” Technically money is created by the Fed during QE, but in equal measure with debt and in such a way that nobody’s equity changes! QE does not change anybody’s equity!

      I don’t think it’s the case that MR doesn’t try to predict anything …for example, MR says that it’s EXTREMELY unlikely that the US won’t be able to pay its bills… at least given that we don’t have any major calamities (hackers erase all the Fed’s computers at the same time Washington DC is obliterated with a terrorist NUKE) … or congress decides to default to teach America a lesson…. or for some other brain dead idea. But basically I think you’ve got it.

  13. As I understand it, when the Fed buys back a bond the bond no longer exists. So there is no “down the road”.

    • Actually, the bond still exists, and the Treasury must pay the Fed the principal back (interest payments are remitted back to the Treasury). Plus the Fed could sell the bond later before it matures.

  14. Midas2, Thanks, Those bonds are effectively retired from the stock of assets circulating in the financial system (true, although perhaps only temporarily), but they have not expired. The Treasury pays interest, and the Fed sends some back minus expenses according to their published accounting. But the Bonds have not gone to maturity yet. That remains “down the road”. I think we are still in that vacuum of speculation.

    • This is not a very realistic example (my link below): I’m making all kinds of simplifying assumptions, but I’m following through to the logical conclusion of what happens when the Fed buys Tsy debt and holds it until maturity, an then more Tsy debt is issued to pay the principal back to the Fed, and then more Tsy debt is issued to continue deficit spending, etc… I tried to conclude at a place which is similar to the start: The overall effect is not really different than just continued deficit spending. In my examples I like to start off with all zero or nearly zero balance sheets for everyone… and given that huge oversimplification, the effect of continued gov deficit spending, whether the Fed buys and hold the Treasury’s to maturity or not, is that the Treasury essentially issues more and more debt instruments and the public acquires more and more. So private sector equity goes up by the same amount that Treasury goes down… and you can always return to a place where there are no deposits present anywhere. So that’s how I like to see it: Fed holding to maturity or not, you can always stop and “sample” the process at a place where the private sector holds an ever growing pile of Tsy debt, and the Tsy likewise has an ever growing Tsy debt liability:

      Hope that helps! Also, you might check out my Example 8 too.

      • Tom, I agree that you have described MR. “the Treasury must pay the Fed the principal back (interest payments are remitted back to the Treasury)” You’re describing how things have been going. But the situation we have now is unprecedented. You must agree that our ultimate central bank has done something that was only discussed in the 1930s. The Fed has built up a balance sheet that is unlike anything the world’s economy has seen before. I’m speculating that at maturity these bonds are written off. Do you think “Treasury must …” is controlled by a law or something? I’m just asking, I have no clue.

        • I’m sure somebody is going to speculate that such a “write off” will cause massive inflation. But most on this board know better. Inflation is caused by a different factor: Borrowed money chasing overbought assets. Way too much credit sold and “printed” by under regulated banks and non-bank institution.

          I speculate that the world’s central banks working in sic can write off the bonds they have amassed from their various sovereigns and NOTHING will happen. Making these “write offs” too fast or too slow will be seen in currency markets and nowhere else.

          Please understand that I am speculating in a vacuum.

          • I rule now is that The Treasury cannot overdraft its TGA (Fed deposit) … since the early 1980s or late 70s I think, and that it must turn to the private sector for funding (purchasing its Treasury’s at auction). Of course these rules could be changed. And I think you are right, if they were, then that wouldn’t necessarily cause inflation.

            But if the rules were not changed, I don’t think it’s necessarily impossible to pay down the debt either. We did after WWII at the debt/GDP ratio was even higher then. There was some coordination, in that I think the Fed capped interest payments on bonds… something like that. Of course after WWII the US was in a unique position in the world, with lots of undamaged cities and industrial capacity, whereas Germany, Japan, Europe in general were not doing as well and weren’t in as strong a position.

              • I still think the present Fed balance sheet is unpresidented. There is no doubt that the debt/GDP ratio of during WWII was very high and that deficit spending had a lot to do with finally bringing us out of the depression. But GDP was quite low, and thus in real terms so was the debt per capita. Finally, the Fed did not buy sovereign debt in the 1940s. It marketed the debt pretty much like now.

                I think what you’re suggesting is: “don’t worry, our kids are on the hook for this but they’ll have plenty of moola by then.”

                • Fed balance sheet: yes, probably true there! Yes, private debt was very low at the end of WWII (probably in part because everyone was still in the save-every-penny mode from 15 years of depression and war, while the government was spending like crazy).

                  I don’t know if I’m suggesting our kids will pay it off fine!… but there are things that can be done to mitigate the effects of paying off the debt, even in a rising interest rate environment. There’s the post-WWII capping idea, but then there’s also the idea that Tsy could just stop issuing long term debt. If 30 year interest rates are high, don’t issue 30 year bonds. Also, the Fed can control any part of the yield curve it wants to: they can effectively set not just the overnight rate, but say the rate on 10-year bonds through OMOs. They’re not doing that now.

                  • I think I got my answer––
                    According to JKH:
                    “In fact, the Fed will decide if and when it wishes to sell back its QE bonds to the private sector as a function of a QE “exit” strategy. Moreover, many of the bonds may end up being matured on the Fed’s balance sheet instead of being sold back. So it will be the Fed’s option to sell in the case of both the occurrence and the timing of any such exit transactions. There is no legal or economic obligation, as is the case with actual repo transactions.”

                    “If the Fed decides to hold these bonds on its balance sheet until they mature then that’s the grand “exit” strategy. The bonds will simply roll off slowly and the Fed’s balance sheet will naturally shrink.

                    “The majority of people out there seem to think the Fed has to “unwind” QE at some point and the fear mongering people on certain websites and representing certain political views like to portray this as some sort of world ending moment.”

                    Thanks Cullen, This is what I have been wondering about because it’s not just websites, so many “main stream” pundits (Fox and cNBC famous Santelli: Collateral Damage of QE Exit), are panicked about this. I’m just trying to reconcile these two roads. It’s not possible because they seem to be travelling in different realities.

  15. I think MR should clarify ist position as follows (just an example):

    QE is outside money printing of reserves (not banknotes, though reserves can be converted in bank notes).

    QE can be done with a) banks, b) private non-bank sector

    In case a) the reserves may never enter the money supply, as the latter depends in this case on banks lending. The only way reserves can enter the Money supply is through bank notes withdrawn by depositors on queues in front of banks / ATMs.

    In case b) the banks will swap the reserves with inside money, so QE directly enters the money supply. But QE has traditionally been a minor source of money supply, as the latter has consisted mostly of inside money. This historic relationship could change though (is changing right now). Although inflation danger is currently low, as QE offsets private sector deleveraging mostly, if the speed of QE accelerates, it could become inflationary. QE has also psychological and redistributory indirect (negative) effects. Etc.

    • InvestorX, I’ve tried to illustrate your case b) here:

      Maybe we’re on the same page, but I can’t quite tell. As you can see, in case b) the reserves grow the bank balance sheet, but not the non-bank balance sheet: the reason is because the reserves cannot go directly from the Fed to the non-bank. Thus the Fed gives the bank reserves and the bank in turn increases the non-bank’s deposit.

      In your case a), only the Fed’s balance sheet grows, as it’s a direct exchange of assets for the bank (thus the banks’ BS doesn’t change size).

      Regarding whether its inflationary or not, I’ve seen it argued both ways and I’m agnostic to that at this point. Certainly I think if the Fed were purchasing the basket of goods in the CPI it couldn’t avoid being inflationary, but it’s not doing that. Does it inflate financial asset prices though? Seems very plausible.

      • “Regarding whether its inflationary or not, I’ve seen it argued both ways and I’m agnostic to that at this point. Certainly I think if the Fed were purchasing the basket of goods in the CPI it couldn’t avoid being inflationary, but it’s not doing that. Does it inflate financial asset prices though? Seems very plausible.”

        As I meant inflationary, I meant a rise in money supply, not in CPI. A rise in money supply can manifest itself in CPI and/or (financial) asset inflation. Obviously QE with the public leads to higher money supply, unless the latter is offset by inside money deleveraging.

        That QE can inflate financial asset prices is not that obvious to me. For example QE did not stop the bear market of 2009, it was the end of mark-to-market. Then it is also not clear which effect of QE allegedly raises asset prices – i) money supply increase; ii) psychological effects/misunderstandings, e.g. TINA / Fed model, (hyper)inflationary hedging/frontrunning; iii) stimulates speculative activity, e.g. banks prefer to give margin debt / speculate in risk assets, then create loans at these low rates / low growth / no creditworthy borrowers situation (kinda the Greenlight founder’s critique; iv) anything else?

      • Actually JKH explains my case b) in his new post. I meant the same.

        -> QE is “money printing”, although it is (historically) a minor source of money printing.

  16. But what about when you have primary dealers *only* buying newly issued government treasuries, because they know they’ll get to immediately offload those to the Fed, thereby enabling the issuing of more new treasuries than otherwise? Isn’t this money creation?