Yes, Virginia, Banks Really Do Create Money Out of Thin Air

So now from the original $10,000, an additional $17,100 has been created, for a total money supply of $27,100. This process of money creation then continues, as Modern Money Mechanics explains, such that

Carried through to theoretical limits, the initial $10,000 of reserves distributed within the banking system gives rise to an expansion of $90,000 in bank credit (loans and investments) and supports a total of $100,000 in new deposits under a 10 percent reserve requirement…. The multiple expansion is possible because the banks as a group are like one large bank in which checks drawn against borrowers’ deposits result in credits to accounts of other depositors, with no net change in total reserves.

All of this naturally means that even if the original $10,000 was actually printed—that is, even if it existed in the form of currency, i.e., Federal Reserve Notes—most of the so-called “money” that now exists in the system is in the form of credit, created out of thin air, backed by nothing, and existing merely as digits in a computer. As the Fed explains:

Currency is something almost everyone uses every day. Therefore, when most people think of money, they think of currency. Contrary to this popular impression, however, transaction deposits are the most significant part of the money stock. People keep enough currency on hand to effect small face-to-face transactions, but they write checks to cover most large expenditures….

Since the most important component of money is transaction deposits, and since these deposits must be supported by reserves, the central bank’s influence over money hinges on its control over the total amount of reserves and the conditions under which banks can obtain them.

Ah, but we’ve been ignoring one additional factor: the interest that is due on the loans. All of the money created by the banks out of thin air must be repaid to the banks plus interest. This includes the initial $10,000. The government didn’t directly borrow this money from the Fed, but through open market operations, it has issued an IOU for $10,000 plus interest (Treasury securities), which was ultimately purchased by the Fed with money created out of thin air; thus, the $10,000 was in effect simply borrowed into existence, at taxpayer expense.

This means that every “dollar” you carry in your wallet or purse, every digital number in your bank account does not, in fact, represent an asset, but a liability. Under the Federal Reserve system, money is debt. Every single dollar was initially borrowed into existence, and the money supply equals the principle on all outstanding loans.

Which leads us to the question: if the principle amount of all loans is paid off—which, remember, is deflationary—there would be no money in circulation, so where does the money come from to pay off the interest? Well, that, too, must be borrowed into existence. This is why the Federal Reserve sets a target rate of inflation; it must perpetually create more and more money out of nothing so the banks’ loans can be repaid plus interest. This is why the dollar has lost 95% of its purchasing power since the Fed was created in 1913. This is unsustainable, of course. It is a Ponzi scheme, a confidence game.

To sum up, our entire monetary system is a debt-based system, in which every dollar in circulation, whether in the form of currency or credit, was borrowed into existence in the first place when the banks created it out of thin air while at the same time charging interest for its use. But while the money created out of thin air by the banks represents nothing and is backed by nothing, the money it must be repaid with represents actual labor and production of the borrower, something of real value. And, of course, if the borrower is unable to repay the loan, the bank may take the collateral on the loan, such as the borrower’s car or house—again, something of real value in return for something of no value. By such means, the bankers under the Federal Reserve system profit by defrauding the public of their real wealth both through the process of fractional reserve lending and through inflation itself and the loss of purchasing power of the dollar it represents, which inflationary monetary policy also has the additional negative consequence of creating the cycle of artificial booms and recessionary busts. (For more on how the Fed created the housing bubble through its inflationary policy of keeping interest rates artificially low, see my book Ron Paul vs. Paul Krugman: Austrian vs. Keynesian economics in the financial crisis.)

Now all of this, of course, is precisely how Murray Rothbard explained that banks create money out of thin air, i.e., engage in legalized counterfeiting, and this is precisely what Ron Paul has tried to educate people about. This magic of fractional reserve banking, this “witchcraft”, is also precisely what Tamny would have his readers believe is merely a “fiction”. It is most unfortunate than in ostensibly trying to correct alleged misconceptions about the banking system by debunking a supposed “myth”, what John Tamny has in fact done is just the opposite. He has done his readers the great disservice of woefully misinforming them about the true nature of fractional reserve banking.

Join Liberty Classroom today and get 3 FREE books!

Jeremy R. Hammond

Jeremy R. Hammond is an award-winning political analyst, editor and publisher of Foreign Policy Journal, and author. His new book is Obstacle to Peace: The US Role in the Israeli-Palestinian Conflict. Read the first chapter FREE at! 

Comments are encouraged, but please respect the rules. Click here for terms of use.

  • Richard Wicks

    > there are those, most notably Rep. Ron Paul, who
    > believe that no reserve requirement is enough.

    It seems that no reserve requirement is enough, because there are no reserves.

    That is why we had TARP and QE1 and QE2. The banks are bankrupt. They exchanged worthless assets on their books for money the Fed creates with debt.

    The entire system is a joke.

  • Tom Brown

    Please examine the following article, and see if you agree:

    His point is that the “money multiplier” is a myth, but not in the sense that John Tamny meant. Scott (the author) says it’s a myth because reserve requirements (which only apply to checking deposits in the US, BTW… not savings… and have NOTHING whatever to do with loans) limit how profitable a loan can be, but not whether the loan can be created in the first place.

    Do you agree?

    • I’m not sure I understand what you are asking. I think it’s this: Do I agree that reserve requirements do not limit whether a loan can be created in the first place? Yes, I agree, but it does limit how much the bank can create out of thin air.

      Krugman writes:

      First of all, any individual bank does, in fact, have to lend out the money it receives in deposits. Bank loan officers can’t just issue checks out of thin air….

      Krugman is wrong. Banks do precisely that; they lend money they create out of thin air, the amount of the loan equaling the amount of the deposit they create on the liabilities side of the balance sheet, as explained above.

  • Tom Brown

    I agree Krugman is wrong. I’m more interested in your views of Scott Fullwiler’s examples: Figure 1, and Figure 2.

    Figure 2 implies that starting with a system with no fiat money (if we take Figure 2 and the assumed presence of a central bank to be a closed system), we arrive at a a situation, after a loan is made, and a deposit transferred between banks, where the central bank was FORCED to create fiat money that Bank B acquires as reserves. This money is called MB, no? Also, there’s this, from Steve Keen, in reference to the following study:

    Kydland & Prescott, Business Cycles: Real Facts and a Monetary Myth, Federal Reserve Bank of Minneapolis Quarterly Review, Spring 1990.

    Here’s an excerpt from Dr. Keen’s blog about this:

    Their empirical conclusion was just the opposite: rather than fiat money being created first and credit money following with a lag, the sequence was reversed: credit money was created first, and fiat money was then created about a year later:

    “There is no evidence that either the monetary base or M1 leads the cycle, although some economists still believe this monetary myth. Both the monetary base and M1 series are generally procyclical and, if anything, the monetary base lags the cycle slightly. (p. 11)

    The difference in the behavior of M1 and M2 suggests that the difference of these aggregates (M2 minus M1) should be considered… The difference of M2 – M1 leads the cycle by even more than M2, with the lead being about three quarters.” (p. 12)

    Thus rather than credit money being created with a lag after government money, the data shows that credit money is created first, up to a year before there are changes in base money. This contradicts the money multiplier model of how credit and debt are created: rather than fiat money being needed to “seed” the credit creation process, credit is created first and then after that, base money changes.

    So what this says to me (Tom speaking again), is that the banks can force the creation of fiat money from the central banks… thus there’s NO effective limit that reserve requirements put on the creation of bank money. Also, isn’t it true that Canada does not have any reserve requirements, thus no limits in Canada, even if the idea of limits was true?

    Thank you so much for taking the time to answer!

    • I don’t see how that could really be the case, since the amount of loans a bank can make (i.e., the amount of money it can create out of thin air) is limited by the amount of reserves and the reserve requirement. I don’t understand the argument at the link you provided that this isn’t so. The argument seems basically to be that banks ignore the reserve requirement altogether. That may be, but I’m not convinced of that. Having something in reserves is just smart banking, since the bank will want to be able to meet any demand for cash, and loaning out beyond the 10% requirement could just end up putting the bank at risk. True, the Fed acts as “lender of last resort” and could bail out any bank that got into trouble that way, and maybe that is essentially what the argument is, that credit creation does not follow the monetary base, but leads it, because the Fed just constantly bails out the banks by running the printing presses, so to speak. Again, that might be the case, but I don’t think so. The Fed’s monetary policy is driven more by just trying to achieve a target rate of inflation and whatever target interest rate. As for the quotes about M1 and M2, I don’t see what it has to do with the question at hand. That paper is about the business cycle, not the creation of credit. It says that M1 does not lead the business cycle. It does not say M1 does not lead the creation of credit. What does that even mean, to say there’s no evidence M1 leads the business cycle? I don’t know. It seems to me to be an attempt to deny that the Fed’s inflationary policy causes the artificial boom (in which case I would think it not credible), but that paper is rather technical and over my head, so I can’t make any sense of what it is really trying to say in that portion Keen quoted, and I don’t have time at the moment to try to decrypt it further.

      Links for anyone else wanting to know what we’re talking about:

      • Tom Brown

        Jeremy, thanks so much for your reply. I am definitely not an expert in this field. It seems to me that I ought to be able to verify whether or not Scott Fullwiler’s ideas are correct or not by asking any bank manager.

        I think I understand your response and skepticism. Let me see if I can argue Scott’s point of view in my own words:

        First of all, in Figure 1, we could assume that Customer 1’s deposit was in a savings account rather than a checking account. As I understand it, reserve requirements do not apply to saving deposits (and of course reserve requirements are ONLY on deposits… not loans). Or we could assume that the account is in Canada or Australia, or some other country with no reserve requirements. So I don’t see a problem with Figure 1. If Customer 1 were to close his savings account and move it to a checking account, then Bank A would have to raise the required reserves. It could do this by borrowing reserves from other banks, borrowing from the Fed, or attracting deposits… (could it also do this by selling shares of stock? I don’t know). In any case it can certainly borrow the reserves. Of course I’m ignoring any capital requirements (Basel accords) on the original loan (as opposed to the deposit) (and I’ll continue to do so, since I don’t understand those!). If the Fed is to maintain a target interest rate, it is compelled to make loans at the Fed funds rate that are demanded… so if there’s absolutely no way for Bank A to raise the reserve requirements in any other similarly priced way (money market), then the Fed must loan it the money (otherwise, another bank would be free to raise the rate that it gets for reserves, and the target rate would be violated). I think this gets to the idea that the central bank can either choose to directly limit the base money supply, or the overnight rate, but it cannot do both. In our system it chooses to control the interest rate.

        In Figure 2, something else is happening… here the deposit is transferred. This doesn’t have anything to do with reserve requirements (we could again assume we’re in Canada or that we’re dealing with a savings deposit). In order to support the payment clearing system (something Scott argues is part of the charter of the Fed), the Fed is again compelled to create the reserve balance that Bank B is credited with. The central bank can require that those funds be replaced in a day, but again, Bank A can accomplish that by borrowing the funds from Bank B, raising capital, attracting deposits, or borrowing from the Fed. Deposits are attractive because the interest rate Bank A must pay for those loans is even smaller than what it pays the Fed or Bank B (what it must pay Bank B or the Fed is presumably the same, because again, in order to defend the targeted overnight rate, the Fed must loan at that rate). Scott doesn’t explicitly say that Bank A could borrow from Bank B to cover it’s reserve deficit, but I take it that he implies that by the reference to borrowing on the “money market.” I like the idea because it makes Figure 2 a tidy closed system.

        You refer to these forced operations at the central bank as bailing out the banks… but if that’s the way the system works, is that really a fair assessment? If there’s an implicit guarantee that you can always borrow the funds needed to cover reserve deficits (after a bank to bank transfer of deposits) or to cover reserve requirements, then perhaps it’s like Scott says: The reserve requirements do not directly limit the quantity of money that can be loaned in any way. All they do is affect the profitability of loans (and thus perhaps indirectly influence the quantity of money loaned).

        In other posts that Dr. Keen has made, he’s also said that in the US the banks are only subject to reserve requirements on what he called “household” demand deposits (which I take to mean checking deposits of private individuals… not businesses). He gave a reference for that which I haven’t verified yet. Also, as I understand it, the five biggest banks do not have to abide by the 10:1 or 12:1 reserve requirements (or whatever they are), but instead have negotiated levels of 30:1 to 50:1 with the regulators… and again those are only for SOME of their deposits. Do you know about this? All of this leads Dr. Keen to state that banks effectively don’t have reserve requirements, even in the US. Here’s the reference for Steve’s assertion about the “household” deposits… with a reference to a table in a report about OECD countries that gives more info (search for the word “household” in the following):

        The case of Canada is especially interesting to me though, if indeed they don’t have any reserve requirements at all on ANY deposits. From the limited description that Scott gave of the Canadian system, it sounds like the private banks explicitly avoid keeping any reserves at the bank of Canada (BOC?) overnight… i.e. the whole purpose of the “reserve” accounts at the BOC is thus to settle accounts during normal business hours. Thus the BOC only has the role of setting overnight interest rates and settling payments. That’s what it sounds like, but I’m not sure I’ve got that correct. I’ve written Scott Fullwiler for further information…I’ll let you know if he responds to my questions.

        I would love to get to the bottom of this, because it has huge implications! If indeed the banks can create credit money out of thin air with effectively NO restrictions on the quantity of money created, that is entirely different from a system wherein the central bank controls the upper bound on the money supply by limiting the “seed money” (base money) to the money multiplier machinery… especially during bubbles, when the demand for credit is high! BTW, I say “upper bound” because if there’s low demand for loans from entities (that the banks are willing to lend to), then the money doesn’t get multiplied (or multiplied by the full amount), and it stays locked up in reserves, or it gets used to purchase treasuries, or it get’s used for some other speculative purpose. I think that’s the situation we have now. Nobody want’s to take on new debt, despite the Fed’s best efforts to re-inflate the bubble.

        • I don’t understand what is meant about reserve requirements affecting the profitability of loans. I also don’t know the details of the reserve requirements. In my view, yes, it’s fair to say the Fed would be “bailing out” a bank by acting as lender of last resort and simply creating the money to bring the bank up to its reserve requirement if the bank was to overextend itself by lending beyond the requirement, because in this manner, the Fed would be keeping banks “solvent”, so to speak (setting aside banks being inherently bankrupt), by essentially robbing the wealth to do so from savers. Running the printing presses (or the digital equivalent) does not create wealth. So this would be a kind of bailout, in my view. Or the Fed might not create the money to help the bank meet its requirement, but simply loan to it from existing reserves at the Fed. I don’t know how that is supposed to work, according to this hypothesis. But, again, how much the Fed inflates I don’t think is driven by banks going out and lending beyond their reserve requirements, but rather by their target inflation and interest rates. My understanding is the way the Fed lays it out in Modern Money Mechanics, that the amount of money that can be created out of thin air is limited by the Fed’s monetary base and the reserve requirement. If a bank needed to borrow from the Fed to meet its reserve requirement, its not a free hand-out from the Fed. It’s a loan, and must be repaid, so I do not see how it is accurate to say the amount of money created out of thin air is not limited by reserve requirements.

          I don’t really know about any of that stuff. What I do know is that banks create “money” of out thin air through fractional reserve lending.

          • Tom Brown

            You wrote: “If a bank needed to borrow from the Fed to meet its reserve requirement, its not a free hand-out from the Fed. It’s a loan, and must be repaid,…”

            I think that’s exactly the point about how reserve requirements don’t directly limit the quantity of funds a bank can loan… they only affect the profitability.

            In other words, the bank is free to make as many loans as it wants, provided it can absorb the reduced spread in interest between what it collects for its loans, and what it must pay to borrow any reserves it doesn’t have. That’s why attracting transfer deposits (either from other banks or from cash) is to the bank’s advantage: the rate it must pay on those deposits is even smaller than what it pays to borrow reserves on the money market, thus the spread is greater, and it’s profits higher.

          • I see. I’m still not convinced it actually works that way, but it is interesting.

  • Tom Brown


    Imagine a hypothetical situation with a single bank, two people, and a single asset, such as a house, and no actual money… only the definition of money that all parties accept. So if person A owns the house and person B wants to buy it, person B, having no money, goes to the bank and requests a loan with the house as collateral. The bank loans him the money, simultaneously creating an loan-asset on it’s balance sheet and a deposit-liability. B buys the house from A, and the bank transfers the deposit-liability from B to A. Now A wants to buy the house back for twice the price, so he too goes to the bank, takes out a loan for the difference between his asset-deposit (the bank’s liability) and the new house price.

    This process can continue indefinitely, inflating the house price, without any base money in the system whatsoever, and with no interference from some central bank.

    Of course I’m assuming no reserve requirements, capital requirements, central bank, or interest paid.

    But interest can be added… suppose the person that most recently sold the house (for a profit), say person A, uses some of his deposit funds to hire the other person for some job. Then B can afford to pay his interest and principal payments for some time.

    Of course this won’t last forever, so we’d need to bring in more people in this economy… all borrowing their funds from the one bank. This economy depends on the growth of debt.

    Of course at some point the whole scheme falls apart when people can no longer afford their debts. They default on their loans, and the bank ends up collecting its collateral (all the goods produced in the society).

    I know this is just an ultra-simple model, but I think it illustrates how banks, on their own, can create a bubble economy by growing debt beyond sustainable levels. In other words, it doesn’t necessarily require a central bank or any base money.

    • I don’t think fractional reserve banks could create bubbles, at least certainly not to the same extent, without a central bank. For one, the cause of the bubble is artificially low interest rates, which sends the signal to investors and entrepreneurs that people are deferring consumption and there is a large pool of capital available to finance long-term projects, so they invest in capital goods or assets. But the capital isn’t really there, so the growth is illusory and not sustainable, and eventually as prices in that sector of the economy rise, projects start going bust and the malinvestment reveals itself.

      Two, without the central bank, the banks would be more limited in how much they could inflate. Rothbard also explains this in The Mystery of Banking. If Bank A issues its own notes, why would anyone have confidence in them? Without legal tender laws and a legal monopoly over the money supply, people would naturally incline towards the use of whatever commodity the market determined as money, like gold. And a bank would have to build confidence by establishing a reputation for sound banking practices and an ability to promptly redeem any demand for gold. The risk of bank runs would keep the fear of God in the banks, and they would tend towards higher, not lower, amounts of reserves. And banks would be more competitive with each other. Each bank, with its limited clientele, would need to be able to make redemption for nonclients, for clients of competing banks. So if Bank A creates credit out of thin air and lends to to person A, and person A buys something from person B, and person B deposits the check in Bank A, there will be no pressure on the bank, because its reserves will remain the same. But if person B is the client of another bank, he might demand redemption from Bank A. Or alternatively he will deposit the check with his bank, Bank B, and Bank B will draw down reserves from Bank A to cover the new demand deposit. Either way, if Bank A can’t meet that demand, it is instantly out of business. Banks would compete against each other and to meet the demand for redemption of their notes, with no central bank ready to bail them out as “lender of last resort”, they would tend to keep reserves higher, not lower. It is the central bank that allows the fractional reserve banks to inflate in a cooperative and coordinated manner. This is the entire purpose of the central bank, to provide “elasticity” of the money supply, as it is called in banker parlance.

      • Tom Brown

        Points taken. However, in the system we have (and perhaps you’d agree… I’m not sure), if Bank A runs a reserve deficit (at the Fed) it can always borrow the reserves to fund the deficit…

      • Tom Brown

        Regarding people accepting the bank money… here’s one argument I’ve heard: the government gives value to fiat money by requiring that tax liabilities be paid in it. If the government also grants a charter to a bank to create deposits exchangeable in fiat notes, then it gives the bank all it needs to have its deposits accepted as money. That’s probably an oversimplified explanation about why people accept fiat money, but I find it interesting.

        Somewhere I read (sorry I can’t be more specific) that British colonialists in Africa had trouble getting the natives to work for them for money, so they charged the natives a tax payable only in notes printed by the colonists. This provided all the incentive necessary to get the natives to work for the Brits! An apocryphal story? Perhaps… if I find the reference, I’ll check it out and come back here and let you know!

  • Tom Brown


    I don’t have time to go over your latest post in detail, but I’ll take a look when I get the chance. Just thought you might be interested in this.. an article from the Federal Reserve about the money multiplier. I made it though the first part, but have not yet tackled the full article:


  • Tom Brown

    Woops! I posted the wrong link. That’s a link to your article, duh!

    Here’s what I meant to post:

    • Thanks, Tom. Seems we’re both wrong to an extent. This paper supports what you were saying about reserve requirements not really acting to limit how much money banks can create out of thin air through fractional reserve lending (I see that clearly now), but also seems to me to show that there is no reverse causal relationship; that is, that an increase in lending does not drive the monetary base. The overall thesis here is one I think is fairly common sense. Apparently, from what I gather here, the textbook idea of the money multiplier is essentially: if the Fed creates $10,000 in reserves, then the fractional reserve banks will increase the money supply to $100,000. Which is obviously a non sequitur and I would think all these smart economists would not have to be told this is a fallacy and there would be no need for such a paper as this to show how that is wrong. What is new to me is this idea that $100,000 seems not to be the theoretical limit of how much banks could expand the money supply to given $10,000 in reserves, since they can fund loans via other liabilities on their balance sheet besides those that require reserves (i.e., other than demand deposits), which I think is the main point you were making, no?

      Of course, when Rothbard was writing, and when Modern Money Mechanics was produced, things were different. This paper notes that prior to ’95, reservable deposits did constitute the largest source of funding, but that has since changed as banks have found other means of finding “external funding”, which the paper seems to define as funding outside of the Fed (i.e., other than reserves). Securities now seem to play a leading role, from what I gather.

      I find it curious that this paper does not note that the Fed is paying interest on excess reserves.

      • Tom Brown

        Thanks for taking the time to really go over that Jeremy… I never did get back to it, but I’ll take your interpretation as correct until I do get the chance to review it myself… and chances are I won’t change my mind, since I think you are an honest guy. It’s nice being able to discuss these matters with a level head.

      • Tom Brown

        Hi Jeremy,

        You wrote “…which I think is the main point you were making, no?”

        Yes, that is the point I was making.

        I just re-read your response, which makes me want to dig back into that Federal reserve paper.

        You also wrote “but that has since changed as banks have found other means of finding “external funding”, which the paper seems to define as funding outside of the Fed (i.e., other than reserves). Securities now seem to play a leading role, from what I gather.”

        Hmmm… I guess I don’t understand that completely, so I’ll take a look.

        BTW, I emailed Scott Fullwiler (author of that piece I originally sent to you: “Krugman’s Flashing Neon Sign”) and he finally (and graciously) got back to me after several weeks with a detailed reply to my questions.

        I was specifically asking about the order of causation… can reserves (and thus base money) be generated through private activity in conjunction with the Fed defending the overnight rate, etc. I realize that your conclusion was “no” after reading the reference I provided. However, Scott gave a qualified “yes.” What he said was that if a Fed ends up loaning money to a bank to cover its reserve requirements or to cover an overdraft on its reserve account, then yes reserves are created. However, the Fed frowns on this activity. They will charge a penalty to a bank for an overdraft, and another penalty to borrow from the discount window to cover it… and yet a steeper penalty for a such a loan with no collateral (generally they like to have such loans collateralized). The same is true for loans to make up reserve deposits. If instead a bank borrows on the money market or in some other way, then reserves are being eliminated somewhere, so that no net reserves are created. He also pointed out that not only does the Fed “frown” on such loans, but the market does as well: taking such loans is a sign that a bank is in trouble.

        This led me to think about what happens when a bank receives interest payments. Is that a “penalty free” means of adding base money in the form of reserves? I wrote Scott back with that question, and hopefully he’ll be good enough to answer (although it could be several weeks before he gets to it!).

        As a simple closed system example, I can add interest payments to Scott’s Figure 1 example. Ignoring capital and reserve requirements, say Bank A makes a $100 loan to Customer 1 at a 50% interest rate to be paid back in three $50 payments, starting with the interest. Further assume that Customer 1 does nothing with his deposit except make payments. Obviously he can only make the 1st two payments before he defaults, since he only has $100 in his account. My question to Scott was, what does the bank’s balance sheet look like immediately before and after the default. I took a stab at answering my own question:

        Right after loan is first made:

        $0 reserves
        $100 loan to Customer 1
        $100 deposit for Customer 1
        $0 Capital

        Before default (after 2nd payment by Customer 1):

        $50 reserve deposit (interest payment received)
        $50 loan to Customer 1
        $0 deposit for Customer 1
        $100 capital (shareholder value)

        After default (after 3rd payment missed):

        $50 reserve deposit
        $0 worthless (defaulted) $50 loan to Customer 1
        $0 deposit for Customer 1
        $50 capital

        I put the bank’s capital under “Liabilities” as is customary since that’s the amount the bank “owes” it’s owners. To calculate “capital,” as I understand it, the following formula is used:

        Capital = Assets – Liabilities

        Where of course the “Liabilities” are assessed prior to adding the “Capital” to the Liabilities side.

        So basically I’m assuming that the profits (in this case the interest) is deposited in the bank’s reserve account. Do you know if this is correct?

        Anyway, if I’m correct, then it seems as if the bank, out of thin air, has changed a situation with $0 of reserves into a situation with $50 of reserves (base money) by the act of taking a profit from its loan.

        I realize that the numbers and the terms of the loan in my example are not realistic, but when you take out a mortgage, it’s true that you pay mostly interest during the first years of the loan… so perhaps it’s not that far from reality.

        Where have I gone wrong?

        • I was specifically asking about the order of causation… can reserves (and thus base money) be generated through private activity in conjunction with the Fed defending the overnight rate, etc. I realize that your conclusion was “no” after reading the reference I provided. However, Scott gave a qualified “yes.” What he said was that if a Fed ends up loaning money to a bank to cover its reserve requirements or to cover an overdraft on its reserve account, then yes reserves are created. However, the Fed frowns on this activity. They will charge a penalty to a bank for an overdraft, and another penalty to borrow from the discount window to cover it… and yet a steeper penalty for a such a loan with no collateral (generally they like to have such loans collateralized). The same is true for loans to make up reserve deposits. If instead a bank borrows on the money market or in some other way, then reserves are being eliminated somewhere, so that no net reserves are created. He also pointed out that not only does the Fed “frown” on such loans, but the market does as well: taking such loans is a sign that a bank is in trouble.

          Without going back and rereading what I said, I think my “no” in this regard was not to answer whether the Fed “can” expand the base money supply to cover banks lending over their reserve requirements, which I agree the answer to is “yes”, but rather was to answer the original question of whether credit creation by fractional reserve banks precipitated money creation by the Fed (banks just lend as much as they want and the Fed just prints up all the money to cover their reserve requirements) or vice versa. I was arguing that it is the latter, that the base money supply and reserve requirements do limit how much money banks can create through fractional reserve lending. The explanation you just provided supports my answer. ;)

          On your balance sheet examples, you define capital as assets minus liabilities, but you also include capital as a liability. So capital (c) = assets (a) – liabilities (l), and I’ll make liabilities other than capital “x”, so that becomes c = a – (x + c). This would seem to me like saying that my personal savings equals my income minus my expenses plus my savings, s = i – (e + s), which doesn’t make sense. That means the more I save, the less I save? This is logically impossible. So say I make $100 and I spend $50 this month. So we have s = 100 – (50 + s). But how can we know what my savings are then? What value do we put in for “s”? It’s undefinable. Plug in an arbitrary number, say $10. So 10 = 100 – (50 + 10), so 10 = 100 – 60, so 10 = 40? Again, logical impossibility. Capital cannot be both capital and a liability, by definition, just as my savings cannot be both savings and an expense. I could say my savings is what I “owe” myself, but I still can’t treat it as an expense, because it makes the word meaningless and leads to the logical paradox wherein the more I save, the less I save.

          • Tom Brown

            Hi Jeremy,

            Thanks again for your reply. I see the paradox you propose, but that is why I also wrote:

            “Where of course the “Liabilities” are assessed prior to adding the “Capital” to the Liabilities side.”

            I thought it was strange to do this at first as well, but I’ve seen several sources now that show that’s how the “capital” is entered on the balance sheet. Sometimes it’s called “shareholder value” or something else. But basically you make a normal balance sheet, but if Assets exceed Liabilities (like they better! or you’re out of business), then you make up the different on the liabilities side under “Capital” or “Shareholder equity” or something like that. Here’s one source:


            I got the same story from writing to Scott Fullwiler (or Cullen Roche… I can’t remember now).

            It kind of makes sense… that’s the shareholder’s money, and both sides of the balance sheet better be equal otherwise it’s not in balance any longer.

          • Oh, sorry. Yes, I see in my haste I had missed where you clarified that. So scratch all that. Going back to your question, then, I guess I don’t understand what you are trying too illustrate with your hypothetical accounting. Where does the borrower get the money to repay the loan? If he took the $100 out and spent it all, to repay the loan, he would have to earn the first $50 payment and pay it back in cash, assuming he doesn’t have another checking account at another bank. And if he paid it back in cash, yes, I suppose that would be added to the bank’s cash reserves on the asset side. But I don’t otherwise see how you get to your second T-account entry from the first, because wouldn’t the balance sheet would have to be balanced before you add “capital” to the “liabilities” side? I’m afraid I’m confused and don’t follow (and can’t take the time right now to read those links, if they would help clarify it for me).

          • Tom Brown

            Here’s another source for the equation:


          • Tom Brown

            No worries, I think I see where I went wrong. It should look like this:

            After 1st payment:
            $100 loan
            Liabilities & Capital
            $50 deposit
            $50 capital

            After 2nd payment (prior to default):
            $50 loan
            Liabilities & Capital
            $0 deposit
            $50 capital

            After defaulting on 3rd payment:
            $0 defaulted $50 loan
            Liabilities & Capital:
            $0 deposit
            $0 capital (Assets – Liabilities = 0)

  • DJC

    We agree, i can’t believe so many people believe banks can create money from nothing. I prefer to use the term “velocity” of money rather than the multiplier effect as it better describes money moving around the economy creating new spending power rather than new money. I address some of the reasons many people believe in this creation of money on JeS in particular the growth in m1 and m2 money supply.

    • Precisely how banks create money from nothing is explained in the article. If you think there is any error in fact or logic therein, you are welcome to point it out.

  • DJC

    I think the fed article is talking about QE and the fact it has not incresed demand through an increase in money supply and spending via the multiplier effect. BTW central banks can and do create money from thin air.

    • BTW central banks can and do create money from thin air.

      Indeed, as do fractional reserve banks.

  • DJC

    They are purely talking about the effect of policy ipon economic activity, not the ability of banks to create money from nothing. Which they can’t. This article does remind us that lowering reserves will not increase base economic activity. I beleive the reason for this is most monetary activity is not in the real economy but rather in financial markets. So new money is mulitplied 100 fold in a day through financial trading markets but little if any trickles to the real economy. Therefore the classic money multiplier no longer seems evident.

    • They are purely talking about the effect of policy ipon economic activity, not the ability of banks to create money from nothing. Which they can’t.

      Yes, the paper does not talk about the ability of banks to create money from nothing, which they can and do.

      • Tom Brown

        I like to think of it as the bank having a license to create +$s and -$s. You could claim that they cancel each other out, and that a net of $0 is created (and if the loan is paid back immediately, that’s essentially true). However, both sides of the balance sheet grow. From the bank’s perspective, the +$s are the loan assets and the -$s are the deposit liabilities. The bank’s +$s grow more rapidly than the -$s (i.e. there’s an interest rate differential, if all goes well with the loan) and that’s how the bank makes its money. My growth rate analogy is like a compound interest situation. From the bank customer’s point of view, things are reversed (the signs change) and the -$s (loans) grow faster than the +$s (deposits). This is the part that’s reminiscent of a Ponzi scheme… eventually the -$s will overwhelm the +$s unless the sovereign can pump some more +$s into the economy (w/o the accompanying -$s), or a trade surplus can be established to bring in more +$s to pay off the debts. In our case the government and the banks took measures to keep the bubble inflating, and thus keep the banks solvent longer than they normally would have been (lowering the loan requirements, and making riskier and riskier loans).

        This is where Hyman Minsky’s three categories of lending appear:

        1) Hedging (the borrowers can pay the interest and the interest)

        2) Speculating (the borrowers can only pay the interest)

        3) Ponzi (the borrowers have to go into further debt just to pay the interest)

        Having watched people I know engage in all three forms of lending as the housing bubble inflated to its peak, I can immediately see the correlation!

        I started off buying a property in 1999 using a traditional “hedge” loan (full doc, 20% down, 30 year fixed). But then I bought another property paying “interest only” in the hopes it would rise (a “speculative” loan). And I witnessed people take on “negative equity” loans wherein they went further into debt just to pay the interest!! Crazy! … and all the while Greenspan and then Bernanke kept lowering the interest rates to make sure the bubble kept inflating.

        I’ve read that in ancient Sumeria, Babylon, and Israel, they had a similar problem. Eventually the economy became so overburdened with debts (to the palace) that when a new king took power, often the first thing he’d do is cancel all the debts to the palace… essentially restarting the economy.

        • Tom Brown

          Whoops! meant to write

          “1) Hedging (the borrowers can pay the principal and the interest)”

  • Luke

    When fractional-reserve banking is combined with the gold standard, it’s true bank lending is constrained by reserve requirements, as the supply of gold is finite. However, when fractional-reserve banking is combined with fiat currency, reserve requirements have no constraints on lending as central banks will always add new reserves to the system, through Open Market Operations, when the banks need them. If the banking system creates loans, then demand for reserves will rise. High reserve demand will increase interbank lending rates. In order to ensure interbank rates remain near the central bank’s target rate, the central bank will buy government bonds from dealer banks in exchange for newly created money. Thus, banks do not lend from reserves. Instead, banks are constrained by equity, that is, solvency.

    I also think it’s a little inaccurate to describe a fractional-reserve bank as a counterfeiter. Counterfeiters create new money for their own immediate use. Banks cannot create new money to deposit into their own accounts. Instead, banks create new money to deposit into a borrowers account, under the condition that money must be transferred, along with interest, to the bank’s account over the term of the loan. Thus, fractional-reserve banking is a time-delayed theft mechanism, whereas counterfeiting is instantaneous theft.

    • If the banking system creates loans, then demand for reserves will rise. High reserve demand will increase interbank lending rates. In order to ensure interbank rates remain near the central bank’s target rate, the central bank will buy government bonds from dealer banks in exchange for newly created money.

      I hadn’t considered that. As for counterfeiting, yes, it is a time-delayed theft mechanism, but counterfeiting is not defined as instantaneous theft, but: “to engage in counterfeiting something of value”, “to make a fraudulent replica of”. So it is counterfeiting.

  • Tom Brown


    Do you know where banks store their profits (i.e. the interest they collect)?

    Do they store profits in their reserve account with the Fed? Do they set up a deposit account for the bank itself? I don’t know how the book keeping works in this case.

    When it comes time to pay their employees, overhead, and disperse dividends, do they make payments from this same account?