Yes Mr. Neoclassical Economist, Banks Really do Create Money

There seems to be some confusion in certain circles as to what banks actually create.  Do they just create debt or do they create “money”.  Money is a tricky term, but let’s go back to basics on the design of the US monetary system.

The monetary system in the USA is designed around private competitive banking (see here for a full explanation of our monetary system’s design).  That is, most of the “money” in our system exists in the form of bank deposits.   When a bank makes a new loan this results in a new deposit in the banking system.  All those digits in your bank account are numbers that some bank entered into the system.  And no, they didn’t need to call the central bank first, check their reserve balances or calculate some silly mythical money multiplier before they issued these deposits.  Banks are never reserve constrained.  So they issue loans almost entirely independent of the government (with the obvious exception of some regulations which obviously don’t work to constrain them that well).   The key to understanding the design of the US monetary system is understanding how banks “rule the roost”.  Banks issue most of the money through a competitive market based (demand) process and the institutional design around this system is in place primarily to support the health of this process.  What do I mean by that?

It helps to quickly understand the two primary forms of money in our monetary system – inside money and outside money.  Inside money is money created inside the banking system.  This is bank deposits (loans create deposits).  Outside money is money issued outside the banking system by the government or the Fed.  This includes notes, coins and reserves.  

First off, our entire monetary system caters to banks.  In fact, the entire Federal Reserve system is designed to cater to banks.  The Fed system brings payment settlement into one place.  So, instead of having 1,000 different banks running around issuing their own notes and trying to settle payments among themselves, we have a Fed system which creates an interbank market where payments can all settle.  This is a lot like having one national bank (but not really because the banks remain privately owned for-profit entities).   The interbank market is where the central bank maintains reserve accounts primarily for the purposes of enacting monetary policy (which is designed to influence the price/demand/supply of bank money) and to ensure proper payment settlement.  So the Fed system is really designed around stabilizing the banking system in various ways even though it’s far from perfect.

Further, the primary purpose of notes and coins is to facilitate inside money.  Most of the transactions in a modern monetary system occur in electronic deposit transfers which settle through the interbank market in reserves.  But you can also draw on your electronic account for notes or coins.  This is a convenient, but quickly becoming extinct form of transaction.  But it’s important to understand that notes and coins exist primarily to facilitate the primary form of money – inside money.

Again, you have to get the order of importance here.  Banks “rule the roost” so to say in the monetary system because they issue the primary form of money which greases the economic engine.  Monetary economists have gotten this all wrong for ages with their money multipliers and arcane beliefs that the government controls the money supply through its various mechanisms that are merely designed to facilitate the existence of inside bank money.  The US system is specifically designed around the private competitive market based banking system.  Yes, the government could, in theory, control the money issuance business.  But it essentially outsources it to private banks.

Make no mistake.  This is a banking centric monetary system where the money supply has been almost entirely privatized.  So the question is – do banks issue money?  They don’t only issue money.  They issue the most important form of money because without inside money, there are no transactions, no payments, and no economy to begin with.


Got a comment or question about this post? Feel free to use the Ask Cullen section, leave a comment in the forum or send me a message on Twitter.

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.

More Posts - Website

Follow Me:

  • Tom Brown

    Do you think Scott Sumner understands this? After reading his two recent posts regarding a proposed new separate “monetary authority” to control the supply of “currency” (by which I’m convinced he means paper bills and coins… and which he mistakes for the “monetary base”) I’m not sure he does. I think his view of reality is very very different than yours!

  • Cullen Roche

    Sumner, by his own admission, does not understand how the banking system works. I think he doesn’t think it’s important to understand it. But then again, most neoclassicals don’t even include banking in most of their models.

  • SS

    Really good summary, Cullen.

  • But What Do I Know?

    Very nice exposition. Now, can you explain whether money disappears (is destroyed) when loans are repaid?

  • Cullen Roche

    In the micro sense, yes, money is destroyed when a loan is repaid. Just like money is created when a bank creates a new loan/deposit. But in the aggregate, the monetary system is an ever expanding supply of loans/deposits as market needs require.

  • Jason

    I like how there isn’t a single empirical source in this entire post. I like how you refer to “inside and outside money”, a strongly neoclassical concept popularized by Blanchard.

    I like how you act as if the money multiplier doesn’t imply banks create money, when in fact it implies exactly the opposite, that banks create many money that is many multiples of the monetary base, by definition (without any kind of calculation whatsoever).

  • Jason

    “When a bank makes a new loan this results in a new deposit in the banking system.”

    Seriously, this is word for word simply a description of the dynamics of the money multiplier.

  • Cullen Roche


    If you prefer papers with citations then read my monetary system paper here.

    The money multiplier is a myth. Don’t take my word for it. Read the Fed’s paper on it.

  • Tom Brown

    So you classify Sumner as a “neo-classical?” So Market Monetarism is just another school of neo-classical?

    I wasn’t sure how to fit them into the scheme of things. I think they reject the IS/LM model which (I understand) most neo-classicists embrace (correct?)… but overall, most of their views are really not that different than traditional monetarists (like John Cochrane?) or neo-Keynesians (like Krugman or Brad DeLong), correct? (the MMers seem to have a soft spot in their hearts for lefty Krugman, despite their differences).

    One of them, however, I think would accept your premise about banks creating money… that’s David Glasner. David seems to have good report with the Market Monetarists (both Sumner and Christensen link to his blog), and I think he promotes NGDP targeting, but I’ve read where David says he basically accepts the “endogenous” view of money. I don’t think he’d be on board with MMTers or even Steve Keen, but perhaps he’d accept the basic MR view.

    I would love to have seen David’s take on Sumner’s articles, but he hasn’t posted for some time.

  • Johnny Evers

    This concept is a bit confusing to a layperson such as myself. When I think of money creation, I might think of a $1,000 bill being printed and released. Or my account being credited. I have $1,000, no string attached.
    But when a bank issues a $1,000 loan, yes it creates $1,000, but it also creates a debit. The bank gives me $1,000, but I have to pay that back, thus destroying the money, so to speak.
    Isn’t it more accurate to say that banks have a time machine in which they are able to bring future money into the present? This works, assuming there is more money in the future (more resources, more production).

  • Pierce Inverarity

    And you’re missing the key element of the bank money multiplier in that it suggests that banks are RESERVE constrained. This is not true. They are capital/equity constrained by definition, and legally constrained sometimes.

  • Cullen Roche

    Market Monetarism is largely an extension of Friedman’s work. Many of the same basic ideas are still there….

  • Jason

    I’ve already read it, most of the citations are of MMTers who themselves just cite other MMters, who occasionally cite a paper from BIS, which in turn is really just citing papers from more MMTers in one massive circle jerk. The few independent sources don’t verify the most controversial of your claims, there are two versions of the money multiplier. One is the idea that banks will always lend a specific multiple of the base money, this is untrue and almost nobody believes this because banks wont lend if there are unworthy customers. The other version is that the money multiplier represents the absolute limit the banking system in aggregate can multiply the monetary base by, by issuing loans, some also argue that in normal times banking loans approach this limit.

    The most controversial of your claim is the strong implication that banks do not need money to lend money, in which case banks shouldn’t need deposits, I should be able to set up a bank right now and lend billions without having to acquire money first; I don’t have to worry about liabilities as long as the people I lend to deposit the money in a different bank. I have no liabilities, no debt, so I can only make money, I can never lose money. That is, unless you contend that when a bank makes a loan to another bank it in fact creates TWO deposits, i.e. one in the other bank and the other in your own bank to draw from, such that when the loan is repaid the deposit is extinguished. But that would seemingly completely contradict your earlier description of banking, what you’re really describing is just the dynamics of the banking system when money is underneath the limit, which most people agree with anyway.

  • Cullen Roche

    The tendency is to think of money as some physical thing. A note, coin or gold bar. But money is first and foremost a medium of exchange. And it can exist in almost any form. A promise, a number in a computer, a gold bar, etc. We need to stop thinking of money as a thing.

    I don’t think it’s necessary to complicate this idea that money and debt are two different things. They are not.

  • Jason

    The money multiplier implies that if a bank has $x in deposits, say an additional $y in equity capital etc, and another $z in other forms of debt financing, it will NOT lend $z+x+y or more so, but will in fact lend slightly less than that so as to cover unexpected losses, demand for withdrawals etc…

  • Cullen Roche

    You obviously haven’t read it because the paper is not an MMT paper. We rejected MMT almost a year ago, in large part, because it doesn’t focus on the reality that the center of the monetary universe is private market based banking.

    When a bank makes a loan it finds reserves after the fact. If it cannot borrow from another bank then it will borrow from the Fed. The Fed must supply the reserves if necessary. There is no central bank choice in that matter. The central bank does not control the money supply. Banks lend first and find reserves later if necessary.

  • Pierce Inverarity

    Yup, there you go. Banks don’t lend deposits.

  • SS

    This is basically what Krugman said when he lost this debate to Keen. He said that customers might remove their deposits in currency in which case the banks can’t balance their assets and liabilities. That’s a piss poor explanation of reality. In reality, the loan creates a new deposit in the banking system and that deposit stays in the system.

  • Jason

    I have read it, I didn’t say it was an MMT piece ITSELF, I read your section on why you disagree with MMT too.

    I dislike the statement “find reserves”, reserves are just an accounting statement, money not leant. If the bank does not have enough MONEY to provide adequate reserve balances then yes it will have to find money elsewhere, usually from the overnight market. But reserves are a tool on the liability side to satisfy withdrawals, they are not relevant to the expansion of assets unless such an expansion would draw so much money so as to seriously jeopardize its reserve position, putting it danger of a bank run (extremely unlikely), or draw money from capital financing putting it below capital requirements.

  • Tom Brown

    For what it’s worth, here’s my take, and perhaps the source of your confusion: Suppose we were on the gold standard and that we had a hard fractional reserve rule (i.e. that the fractional reserve rule applied to ALL deposits… demand, time, etc.). Then we’d have actual “fractional reserve” banking like in the old gold standard days. If the amount of gold was G and the reserve fraction was R (R between 0 and 1), then the MAXIMUM amount of money that could be loaned out would be G/R (that is if the same G amount of gold were re-loaned out an INFINITE number of times). The actual amount loaned out, of course, would be less than this. So you could say the amount of money was somewhere between G and up to (but not including) G/R. On the interval [G, G/R) in mathematical terms.

    The thing is, we really don’t have “fraction reserve” banking anymore because although there are (in the US) reserve requirements (on household demand deposits only), banks can obtain the reserves AFTER they extend credit. Also we no longer have a fixed amount of G (base money) in our fiat/floating exchange rate system (i.e. we’re no longer on the gold standard). Thus that’s why the “money multiplier is a myth.”

    In expansionary credit times, base money is created by the Fed in RESPONSE to what private banks and their customers do. The Fed defends an overnight interest rate, NOT the amount of “base money.” The Fed creates and destroys base money (paper and coin currency and bank reserves) as part of its mechanism to defend the overnight rate that it’s targeted. That’s why the money multiplier is a myth. We don’t really have “fractional reserve” banking anymore.

    We are now in a different time when the Fed is creating reserves (and trading them for bonds) in an effort to encourage lending, so reserves are no longer lagging credit creation. But still, those excess reserves on bank balance sheets are not “multiplied up” by any money multiplier. That wouldn’t even be the case if we were back on the gold standard with hard reserve requirements! Why? Because the “money multiplier” only specifies an upper bound… the actual amount of money will be less, especially if there’s not a lot of demand for credit. This is the “pushing on a string” scenario that some people refer to.

    So the money multiplier ever only specified an upper bound on the total amount of the money supply, and that only “kind of” worked with hard reserve requirements and a gold standard. I say “kind of” because the definition of the money supply is hard to pin down (there’s M0, MB, M1, M2, M3, M4). Look up “money supply” in Wikipedia for a nice summary.

  • Jason

    I’m not saying that in the slightest. I’m not talking about customers. I’m not talking about money.

  • Cullen Roche

    Are you saying that a loan does not create money? You seem to be having the same issue that neoclassicals have by assuming that the only real form of “money” is outside money….

  • Jason

    Banks lend money, money comes from deposits, capital, or other sources of debt financing.

    At best MMTers and associates have managed to argue that there is a non linear time aspect (I have not ever ever ever ever ever seen a reliable empirical source, such as testimonial from bankers involved in this explicitly or inquiries/empirical research into the workings of a specific bank or bank system for this though), in which banks can extend their asset base first without debiting any accounts and waiting until the next audit such that it gives them time to find the sources or financing after the fact, but this REALLY isn’t significant and REALLY doesn’t actually change much at all.

  • Jason

    Oops, I meant to say currency there, not money.

  • Pierce Inverarity

    Nice summary Tom, thanks for that.

  • Jason

    I’m assuming the opposite, even the most hardcore neoclassical has to assume a loan creates money, otherwise loans would not be able to ‘multiply’ the money supply via the money multiplier.

  • Pierce Inverarity

    Nope, it’s not true. Banks lend when there is demand for loans. They DON’T lend money they have on hand. They can lend far in excess of the money they have on hand. That’s how banks ended up with 40 to 1 asset+liabilities to capital ratios right before the crisis.

    And please, read the Fed paper Cullen cited. That’s your credible source.

  • Pierce Inverarity

    Tom addresses your confusion below very well.

  • Jason

    ” Banks lend when there is demand for loans. ”

    I JUST said that earlier: “banks wont lend if there are unworthy customers.” That doesn’t contradict what I’m saying in anyway.

    Here’s also what I’m not saying, I’m not saying assets must be below the value of liabilities, that’s crazy, the value of assets MUST be above liabilities to be able to make a return. What you seem to be referring to is leverage ratios, a leverage ratio is usually calculated as the value of its risk weighted assets divided by (usually) Basel 2 defined capital. That has NOTHING to do with what I’m saying either, assets will always be many multiplies higher than capital. Capital is largely only held by banks due to regulations anyway.

  • SS

    A bank needs capital to exist. No one is denying that. But it doesn’t need reserves to exist. Reserves are an asset of the bank. If the Fed relaxed reserve requirements to 0 as Canada does, it wouldn’t matter one bit to the way banks lend money. Banks would still lend by issuing deposits. There’s no reserve constraint. It’s a capital constraint. Cullen wouldn’t disagree with you that a bank without any capital is incapable of making loans.

  • Tom Brown

    I’ve been trying to sum up MM views in my head. I think this is perhaps accurate: MMers think that if the intention of the Fed (or the new “monetary authority”) is made known that they will do what’s required through controlling the amount of “base money” (and thus the money supply) to keep NGDP growing at some nominal rate, then the market will operate as if there is NO money (i.e. it will essentially be a pure barter system) and that will resolve most of our problems (i.e. then the market will be unimpeded in achieving “equilibrium”… the fabled neo-classical Shangri-La of a well functioning market). This will in essence remove the “money illusion” (The name of Sumner’s blog), and the market will behave in an ideal way. The neo-classicals don’t believe that banks, money, or (private) debt matter (just like Keen always states about their views), and if the Fed would only target NGDP instead of inflation, then that ideal would become reality. What do you think? Does that sound like a good MM summary?

  • Jason

    In other words, having a huge leverage ratio is not in anyway equal to not lending money they acquire. I can make myself have a gigantic leverage ratio if I borrow a huge amount from the bank, get a few people to invest in me, and then invest/lend all of that to others. I could easily make myself have a leverage ratio of 90000:1, and if I am lucky with my investment, could have a ratio of 40:1 of assets to capital+debt.

  • Jason

    Banks in Canada still have reserves, they just decide the amount of reserves that is appropriate themselves rather than relying on a regulatory requirement.

  • Pierce Inverarity

    Dude, calm down. We’re trying to have a debate here, not a shouting match.

    And I’m not putting words in your mouth:
    “The money multiplier implies that if a bank has $x in deposits, say an additional $y in equity capital etc, and another $z in other forms of debt financing, it will NOT lend $z+x+y or more so, but will in fact lend slightly less than that so as to cover unexpected losses, demand for withdrawals etc…”

    Banks lend FAR MORE than these totals, that’s the point of the leverage comment. Banks do not go out and get deposits then lend them. Talk to any banker and they’ll agree. They may not have thought about it, but they’ll agree. Also, the Fed agrees. Kind of hard to argue with the conclusions of the central bank…

  • SS

    Bad analogy. You can’t lend money like a bank does. A bank creates new deposits which create new purchasing power. If you lend someone money you have to physically give them dollars. And if you decide to create those dollars then you’re breaking the law. So no, you can’t leverage yourself in the same way a bank can. A bank is legally allowed to issue new purchasing power denominated in dollars.

  • Pierce Inverarity

    But banks aren’t borrowing to get those ratios…they’re making loans (and deposits) to get to that point in response to demand. My father-in-law is the CEO of a community bank and he and I have discussed this extensively. He does not go out and get deposits before they make loans, nor does he go to the debt markets or equity markets.

  • Jason

    If they’re being forced to “make deposits” then at best you’re describing the non linear time aspect I’m talking about, which isn’t actually that significant. Do you have a source for banks creating 40 times more new loans than capital+deposits+all other debt, note I am not asking for ASSETS being 40 times greater because asset values grow, I’m talking about the initial loan size. Even still asset values being 40 times greater than all debt, capital and deposits would be absolutely remarkable profit of what, 400%? Usually banks make about 1% in profit…

  • Jason

    I don’t have to physically give them dollars, I can electronically transfer the money. The point was to demonstrate that a bank having a high leverage ratio does not, in anyway, imply they are extending credit without acquiring money, since it can EASILY be done whilst still having to acquire the money in the first place. It wasn’t a disproof of MMT, only a disproof of his ‘proof’.

  • Pierce Inverarity

    You’re getting this wrong because you keep thinking they’re getting the deposits first then lending them out. They don’t have to do that. See SS below. Who’s saying they’re being forced to make deposits? Deposits (ex reserves) are a result of loan creation.

    Profit isn’t calculated off of equity or capital at the corporation level. For banks, it’s all about the spread between what the assets are earning versus what they’re paying out for liabilities. Hence the 1% you’re talking about.

  • SS

    Transfer them electronic credits, give them physical dollars, so what? There’s no difference. The point is, banks don’t have to do that. A bank doesn’t obtain deposits before it makes a loan.

    The Fed paper you refuse to read or acknowledge says:

    ” According to the standard multiplier theory, an increase in bank lending is associated with an increase in demand deposits. The data as discussed below do not reflect any such link.”

  • Jason

    For reference, the total liabilities of the banking sector in 2011 was 11,053.9 billion, the total assets was 12,443.5, that makes assets 1.12 times greater than liabilities, not 40, and this is NOT including capital.

  • Colin, S.Toe

    Hoping to get Beowulf’s take on this system, and whether he has thought of a better alternative.

  • Jason

    Dude, even the value of assets (which is going to exceed the amount initially loaned out as asset values grow) of banks in 2011 only exceeds the value of liabilities by .12, and this is not including capital (which could well make up that .12), which makes your statement unambiguously empirically false.

  • Pierce Inverarity

    Assets = Liabilities + Equity. Accounting leverage ratios are calculated by Assets/Equity.

  • Jason

    SS, nothing I’m saying here implies that an increase in lending MUST increase deposits. It will only increase deposits if they don’t already have the money (which they often do) and can’t get funding from elsewhere (other sources of debt, the fed, or from capital), this condition is rarely met. What the Fed is attacking is the idea that banks will always lend as much as they possibly can, which implies that all other sources of funding would have been exhausted, this is not true.

  • Pierce Inverarity

    you’re calculating leverage incorrectly. and, From the Fed paper (which, really, do I have to read it for you?):

    “For perspective, M2 averaged about $7¼ trillion in 2007. In contrast, reservable deposits were about $600 billion, or about 8 percent of M2. Moreover, bank loans for 2007 were about $6¼ trillion. This simple comparison suggests that reservable deposits are in no way sufficient to fund bank lending. Indeed, if we consider the fact that reserve balances held at the Federal Reserve were about $15 billion and required reserves were about $43 billion, the tight link drawn in the textbook transmission mechanism from reserves to money and bank lending seems all the more tenuous.”

  • Jason

    Actually, leverage ratios are calculated by Assets/whatever the fuck definition of capital you choose, which varies considerably, but is usually Basel 2 capital, and is RARELY just common equity capital. But this is completely irrelevant because YOU’RE saying that Assets exceed deposits+capital+all other sources of debt by multiples of 40, NOT JUST CAPITAL.

  • Jason

    Please stop talking, you don’t have the slightest clue at all as to what you’re talking about. I did NOT SAY ANYWHERE that .12 was the leverage ratio, I already explained to you how to measure the leverage ratio here: “, a leverage ratio is usually calculated as the value of its risk weighted assets divided by (usually) Basel 2 defined capital. That has NOTHING to do with what I’m saying”

    I did not anywhere say that 1.12 was the leverage ratio, I was saying the exact opposite, that you were confusing the leverage ratio for the banking sector residual, YOU said that assets exceed liabilities+capital+all other sources of debt by many many multiples, I’m telling YOU that you’re confusing that with the leverage ratio that measures something completely different, and that what you’re talking about, the bank residual, is .12.

    Secondly, “reservable deposits” are NOT THE SAME as all deposits, all liabilities, reservable deposits are only a small proportion of total deposits, and further deposits. Sorry but my statistics do not lie, see here:

  • scharfy

    If you get this one simple fact you understand why private credit creation is a larger indication of potential inflation than fiscal deficits.

    The tipping point for me was when I realized that many banks call their underwriters a “loan ORIGINATION” department.

    Origination in the rawest sense of the word.

    When you get a mortgage they print that money and its your job to earn that money back in the economy, pay off the loan, (destroying the money) and complete the cycle of life.

    This is the nuts and guts of modern banking.




    Important post, IMO

  • Cullen Roche


    I don’t really know. We didn’t put MR together because we wanted to move the policy needle as much as we were trying to get at a more factual operational description of the monetary system. Given the state of the US economy, I think it’s clear that the monetary system has some flaws or at least it’s being mishandled. I’ve argued that private competitive banking is not a problem so long as its overseen properly. Personally, I have a very hard time imagining the govt controlling the entire lending process. Govt as a facilitator and stabilizing mechanism is fine so long as it’s done properly in both reducing the odds of the boom that leads to the bust. I think that requires greater oversight of the ability of these banks to issue debt.

    I am sure Beo has his own thoughts on how best to benefit from the MR understandings….

  • Jason

    Also, if you’re saying that “Assets = Liabilities + Equity” (this is untrue by the way), then you’re completely contradicting yourself. Bank assets are all investments and loans outstanding made by the bank, if these must equal liabilities + equity then that implies assets (including loans) can never exceed deposits + (equity capital) + all other sources of debt financing, which is complete contradiction to what you were saying earlier, that assets exceed this 40 times over.

  • Pierce Inverarity

    I’m sorry but what??? “Also, if you’re saying that “Assets = Liabilities + Equity” (this is untrue by the way)” that’s the accounting definition of a balance sheet!

  • Cullen Roche

    Jason, why don’t you read the Fed paper first and then tell us where it’s wrong. Thanks.

  • Jason

    It’s true in a mathematical sense (and true for a bank if you completely contradict more basic versions of MMT entirely by the way), i.e. the value of assets must equal the value of liabilities plus equity capital (net worth), if that’s what you’re talking about (i.e. not defining assets by liabilities and equity but relating them mathematically)then fine. However if that is the case, then it is absolutely impossible for assets (loans etc..) to exceed liabilities and capital by 40 times.

  • Jason

    Cullen, I will soon if it really disagrees with what I’m saying, but I’m not sure it does.

  • Jason

    Specifically, if it’s saying that more reserves = more lending is false, then I completely agree, but this is only disagreeing with first version of them money multiplier that I was alluding to earlier and which almost everyone thinks is wrong anyway.

  • Johnny Evers

    Government debt, yes, is that a T-bond is a promise to pay, and we know the government has a printing press.
    Banking debt, yes, to the extent the Fed will buy the asset.
    A lot of banking debt we would see as worthless if we were outside the system and didn’t have a personal stake in the emperor wearing clothes. We want it to have value, so it does.
    Debt can ‘act’ as money in financial equations and in most circumstances (I have bonds in my portfolio and I count them as money), so long as the promise to pay is real. Once that disappears, it has no value. So does that make debt less than money? Or less ‘money’ than as asset which is more likely to retain its value?

  • Cullen Roche

    That’s what the money multiplier implies….more reserves = more lending. Ie, the central bank directly controls the available supply of loans. We’re saying that’s wrong so if you’re agreeing then I don’t know what it is that you disagree with….

  • Pierce Inverarity

    Cullen, is it fair to say that Reserves (not Required Reserves) are Outside/Government money?

  • Cullen Roche

    Reserves are definitely outside money. They’re supplied by the Fed and only exist for the purpose of meeting requirements that go along with the purpose of the Fed system’s existence. If there was no Fed system (think one national bank) there would be no need for reserves.

  • Jason

    That’s what the basic version implies yes, and I agree it’s wrong (so do most neoclassical by the way, at least during a recession). What I’m disagreeing with is the idea that banks create money out of thin air which lots of people imply, rather than creating money via double entry accounting. Which means I find the concept that banks do not have to acquire money before they lend money dubious when speaking statically but defensible in the MMT framework where this is non-linear across time whereby banks can attain the funds after lending rather than before or at the same time, but I find it completely unsupported when taking the total time period between bank audits into account and treating it as one point in time. This also implies that base money has a role in determining the total amount of money, but only in so far as it can impose a hard limit on the total (which would be x times the monetary base, where it could be many multiples, maybe even close to 100 times) but not controlling the amount of money whenever the amount is under this hard limit (which it is much of the time).

  • john

    Confused about a few things. ( sorry if these are off topic, I have maxed out my brain on reading for the day, so I have not gotten to actually read the above article, but I will later)

    If less people are working and wages are stagnant to declining, then why are prices rising for things people need? No such thing as supply and demand anymore? Or is this all pass through from higher energy costs? If so, are energy costs higher due to mal-investment?

    If the FED buys financial products, including bonds on the secondary market, how do the proceeds of these transactions not enter the system and cause inflation?

    Deficit spending is a direct injection of new currency into the system. How much deficit spending can a system handle until it destroys the value of the currency faster than productivity and wage increases can keep pace?

    If a bank cannot force people to borrow, then what is the net benefit of larger and larger reserves? Assuming banks do not lend reserves anyhow. I assume bank reserves have more to do with balancing out the books as more and more loans get defaulted on and the banks would need to find more deposits to offset their losses. Since fewer people work, there are fewer deposits, thus the fed fills the hole in the reserve ratios.

  • Colin, S.Toe

    Thanks, Cullen. Your MR accounts have been very helpful in explaining how the system works.

    I also have a “hard time imagining the govt controlling the entire lending process”, but I don’t know enough to identify realistic alternatives to either this or the current system.

    With his combination of knowledge and originality, I’me wondering whether Beowulf has done so.

  • Cullen Roche

    This isn’t MMT. It’s the basics of modern banking and endogenous money.

  • Jason

    I know you’re not MMT but MMT is where I first saw that particular argument advocated and surely where you found out about it first given that you’re a former MMTer so I label it as the ‘MMT argument’. Either way, I haven’t seen a reliable source for it yet.

  • Cullen Roche

    Google endogenous money. It’s not really modern at all. It just took a few decades for central bankers to realize that they aren’t the center of the monetary system. And no, I wasn’t introduced to banking by MMT. I actually learned much of this from reading Keen’s blog back in the day before I ever heard of MMT. First time I ever saw MMT references was on his site and then later on Pragcap when Warren started commenting on occasion. MMT mainly just adds the govt understandings on top of the endogenous money understandings.

  • Jason

    I’m well aware of endogenous money, endogenous money is not as extreme as the claim I’m talking about at all in all cases I have seen except endogenous money theorists who also happen to be MMTers/Neo-Charterlists. But this is really just semantics detracting from the real point. To say banks create money is misleading unless you’re specifically talking about double entry accounting (i.e. adding both the deposit and the loan to the money supply), and even then it’s misleading to refer to that as creating money, when it’s really just increasing circulation and velocity.

    To be honest I disagree with some aspects of this article, but I find it to be far more convincing than everything I have ever read from people like Steve Keen and Neo-Charterlists:

  • Jason


    If you take out a loan the bank is not lending you its deposits.”

    Lending deposits sounds misleading because some people might interpret that as literally reducing the amount of deposits or liabilities to make that loan. What a bank does is lend MONEY, that money came from a combination of deposits, capital, and other sources of financing (borrowings and money from the fed). Even if it only has deposits, when a bank lends money, the deposits are not reduced and depositors still retain all their money, hence money is ‘created’ via double entry accounting, but it is misleading to describe this process as money creation.

    “It is creating a new deposit”


    “When a bank makes a loan it debits the Loans Receivable account on its books. To balance this transaction it will create a new liability in the name of the borrower. This loan will create a deposit somewhere else in the banking system (possibly at the same bank) that will cause this new bank to also account for its new liability (the deposit) and change in reserves at the Fed.”

    Where in this process is this bank crediting without debiting? If they’re not than this doesn’t contradict me. Furthermore, do you have a source for any of this?

    “Are you saying that a loan does not create a new deposit?”


  • Jason

    And if banks don’t lend money from deposits then it is absolutely 100% impossible for bank runs to exist, even though they do happen. All of the depositor’s money is just sitting at the bank, since it is not lent, thus depositors are never in danger of losing their money even if the fed and other banks refuse to lend money to this bank because its all there for withdrawal. This implies that we actually live in a full reserve banking system.

  • Cullen Roche


    The fed paper, which you refuse to read, describes all of this. As do countless papers on endogenous money creation.

    You claim you don’t contradict yourself, but I can’t, for the life of me, see how this isn’t a contradiction:

    “money is ‘created’ via double entry accounting, but it is misleading to describe this process as money creation.”

  • Cullen Roche

    If a bank can’t borrow from the central bank or doesn’t have the cash to meet its withdrawals then it’s insolvent. That’s a capital constraint. Not a reserve constraint.

  • Jason

    “The fed paper, which you refuse to read, describes all of this. As do countless papers on endogenous money creation.”

    I’ll read it soon (however the abstract and a skim reading seems to suggest it does nothing of the sort), but I’ve already read countless other papers which all systematically fail to have PRIMARY sources, they just source each other.

    “You claim you don’t contradict yourself, but I can’t, for the life of me, see how this isn’t a contradiction:

    “money is ‘created’ via double entry accounting, but it is misleading to describe this process as money creation.””

    I put created in scare-quotes for a reason, I’m just using it to signify what people call this process, I’m still saying this is misleading, and entirely separate from another claim, the idea that banks can create money without deposits, capital/equity or other sources of debt, i.e. crediting without debiting. Double-entry accounting money “creation” (again, scare quotes) is not the same as it’s easily possible even without electronic currency but only hard cash, it’s easily possible just via the normal fractional reserve banking and is not in contradiction to neoclassical economics at all.

  • Cullen Roche

    No one ever said banks can create loans without capital. In fact, I’ve repeatedly said the opposite in this thread. Banks are capital constrained. Not reserve constrained.

  • Jason

    “If a bank can’t borrow from the central bank or doesn’t have the cash to meet its withdrawals then it’s insolvent. That’s a capital constraint. Not a reserve constraint.”

    Let me clarify then, if a bank doesn’t use money attained from deposit accounts (incidentally, if this is so then deposits are nothing but an inconvenience and there is absolutely no reason at all that banks would seek deposits) then the bank can never worry about a bank runs in the sense of not having enough money to satisfy withdrawals/transfers to another bank. It MIGHT have to worry about not having enough of the right form of money (e.g. physical cash when there is a high demand for cash withdrawals), but that’s a separate matter and no bank today would need to worry about that.

  • Jason

    “No one ever said banks can create loans without capital. In fact, I’ve repeatedly said the opposite in this thread. Banks are capital constrained. Not reserve constrained.”

    Are you defining capital differently to me? Do think any kind of idle money is capital? To me capital is defined as such based on the form of acquisition, e.g. money through equity financing (usually a small proportion of total assets) is capital, money attained via borrowing or deposits is liabilities.

  • Cullen Roche

    Of course an insolvent bank could have runs. If it’s insolvent then it can’t obtain reserves or sell assets to meet cash withdrawals. Again, this is a capital constraint. You keep proving that banks are capital constrained and not reserve constrained. We’re just going in circles here….

  • Cullen Roche

    I didn’t know there were so many definitions of “capital”. :-)

  • Jason

    “Of course an insolvent bank could have runs. If it’s insolvent then it can’t obtain reserves or sell assets to meet cash withdrawals. Again, this is a capital constraint.”

    I think we’re talking past each other. How in this scenario is the bank insolvent? It’s the opposite, it has all the deposit money, deposits are the debt, thus it has all the money to satisfy its debt, that makes it fully solvent, not insolvent. And nothing it can do, unless it starts using some of that money, can make it insolvent.

  • Cullen Roche

    1. Banks are more complex than loans and deposits.

    2. The banking system is not just one big bank.

  • Jason

    “I didn’t know there were so many definitions of “capital”. :-)”

    Yeah, defining bank capital is a huge can of worms with lots of disagreement, and it’s being redefined again with Basel 3. Just to clarify, you are defining idle money as capital? Because I think many economists don’t, they only define it as such if it comes from a source of financing that isn’t a liability, this could be where all the disagreement stems from, confusion over the use of the term capital.

  • SS

    Basel 3 just redefines different forms of capital. You’re the one who said capital does not equal assets minus liabilities, which is one of the first things anyone learns in economics class.

  • Jason

    No, I said that the idea that assets = liabilities + equity is untrue, as there are other sources of capital than just equity. Furthermore, I maintain that’s not what DEFINES capital, it’s just a mathematical equation that must hold (unless you’re some of the commenter’s here, who maintain assets can exceed liabilities + capital 40 fold).

  • Jason

    In a sense we’re saying precisely the same thing SS, I’m saying assets extended only via equity or other sources of financing is capital, while if its extended via money from liabilities then it’s not counted as capital. That is equivalent to saying capital equals assets minus liabilities.

  • Pierce Inverarity

    You’re completely misrepresenting my position and failing basic algebra:

    SS “You’re the one who said capital does not equal assets minus liabilities, which is one of the first things anyone learns in economics class.”

    You: “No, I said that the idea that assets = liabilities + equity is untrue, as there are other sources of capital than just equity. ”

    You’re contradicting yourself left and right here by not maintaining a consistent vocabulary. The rest of us are all speaking the same language of basic accounting.

    Of course capital can be more than equity. We all understand that equity falls on the bottom of the capital structure. But capital OTHER THAN equity means you are LEVERAGING your equity. That’s where I was talking about banks having leverage ratios of greater than 40 to 1.

    “At the end of 2007, the bankruptcy examiner concluded, Lehman’s real leverage ratio was 17.8 — meaning it had $17.80 in assets for every dollar of equity. It reported a ratio of 16.1.

    By the end of June 2008 — Lehman’s last public balance sheet — it was hiding $50 billion of debt that way, enabling it to appear to be reducing its leverage far more than it was. When investors asked how it was doing that, Lehman officials chose not to explain what was actually happening.”

    Read here (if you can be bothered)

  • Chris

    Are you saying the Federal Reserve is a for-profit bank? If so I disagree. Dont the profits on any Open market ops go to the US treasury? and isnt the Fed more of an alternative entity for private banks to place monies as opposed to lending it among other banks? The latter argument may be less coherent/comprehensive.

  • Tom Brown

    Jason, I read most of your comments and your reference:

    And I think this is where the confusion lies: The author of your reference describes a fractional reserve system. He argues that in a fractional reserve system the re-loaning of “base money” does not increase the amount of base money. I agree. But that’s not our system.

    In a true fractional reserve system, if the central bank were to inject $G into the system (say by perhaps paying government contractors $G), then eventually (if demand for credit warranted it), those $G could be multiplied up, by at most, 1/R where R is the fraction of funds that are required (R lies on [0, 1]) to be kept in the banks’ reserve accounts for each deposit. So an infusion of $G would, over some amount of time, multiply up to a value approaching G/R (given that a demand for credit existed that the banks were willing to satisfy). The amount of “base money” would always stay at $G after it was injected into the economy (I’m assuming here the initial amount of money, prior to the $G injection, was $0). For simplicity I’m ignoring paper bills and coin currency (assume an all electronic system), though that’s not necessary for my argument.

    That’s not our system. In our system, the central bank creates or destroys “base money” (in the electronic case, only bank reserves) at will to:

    1) Keep the centralized payment system of the banks running smoothly
    2) To support the demand for reserves
    3) To defend the overnight interest rate that the central bank sets

    Point 3) is crucial. Suppose the central bank did not attempt to defend the overnight interest rate it sets and instead attempted to control the amount of base money. Let’s say that $G of base money existed, and that was the Fed’s target amount, so it would not create (or destroy) another dollar.

    Say customer A had taken a loan for $X from Bank 1, and left their deposit amount there at Bank 1. Bank 1 would have had to obtain reserves (say 10% of X) to meet the reserve requirements. Say that’s Bank 1’s only customer and 10% of X is it’s only reserve holdings (I’m ignoring capital requirements). Now customer A transfers his deposit to the only other bank, Bank 2 (I’m assuming a two bank world for simplicity). Now Bank 1 is in a jam. It has to come up with the other 90% of customer 1’s deposit to transfer to Bank 2. In such a situation it has to turn to some source other than the Fed to obtain that money, perhaps even Bank 2. The interest rate of those funds will be determined by the market, and the Fed will have NO SAY over it!

    Instead what happens is that the Fed stands ready as a lender of last resort at the overnight interest rate it targeted. Any other potential lender knows this, so they know they can’t charge interest (much) above that amount. So whether or not Bank A ultimately borrows directly from the Fed or from another source to obtain the reserves it must transfer to Bank 2 doesn’t matter: the rate will be the target rate the Fed has set, because the Fed ALWAYS stands ready to defend that rate (and thus always stands ready to provide the BASE MONEY required!).

    Even if we assume that Bank 2 starts off with a clean balance sheet, the Fed ultimately only has to create 10% of X in new base money to defend it’s rate (since in all likelihood, in order to minimize penalties from the Fed, Bank 1 will borrow back from Bank 2 [after the deposit transfer] 90% of X to cover its reserve account overdraft… leaving only a 10% of X deficit it needs to borrow from the Fed).

    The point here is that we don’t really have a fractional reserve system. We have reserve requirements on a fiat money system with a floating exchange rate, and a Fed which targets and does what’s required to defend an overnight reserve borrowing rate. Because the Fed chooses to defend its overnight rate it NECESSARILY gives up its ability to control the base money supply.

    Also, as you’d expect in such a system, we see that the amount of credit money (M2) does not lag MB (the base money). In other words, if we truly had a fractional reserve system (in times of high demand for credit) you’d expect to see any change in MB FOLLOWED… perhaps with a rather LARGE time lag, by an increase in M2 as those fresh MB funds were loaned out over and over again… and *perhaps* eventually M2 would start to approach the new MB/R asymptote. That’s NOT what happens. Instead we see M2 out ahead of MB/R, pulling MB along (again, in times of high demand for credit). Check out the article “Roving Cavaliers of Credit” at Steve Keen’s, and the references he provides there which show this to be the case.

    In times of low demand for credit, in either the system we have or a fractional reserve system, the Fed creating reserves and purchasing bonds with those reserves, does not create much of an increase in M2.

    Another great description of this mechanism is Scott Fullwiler’s article, “Krugman’s Flashing Neon Sign.” He gives concrete simple examples with very simple balance sheets.

  • Jason

    Okay, I’ve found the source of the confusion, you said:

    “That’s how banks ended up with 40 to 1 asset+liabilities to capital ratios right before the crisis.”

    I misread you, my bad, I somehow read that as 40 to 1 assets to liabilities+capital, which of course would be insane and nonsensical, instead of 40 to 1 assets+liabilities to capital (by the way, leverage ratios are actually just (risk-weighted) assets to capital). Still, that’s a straw-man, that would only address me if I somehow claimed that for every $x they extend in loans/investments they hold $x in idle money. I never said that in the slightest, and if you think I did you misunderstood me.

  • Jason

    In sum, my point stands, a bank lending far more than it has in capital is not creating money in the sense that economists mean (crediting without debiting), only if it lends more than it has in debt/liabilities + capital is it doing that.

  • Cullen Roche

    No, the private banks are for-profit. The Fed exists as a support mechanism to the banking system.

  • Cullen Roche

    Yes, Fullwiler is excellent on this stuff.

  • Jason

    Tom Brown, I agree with you mostly but I don’t see where you’re contradicting me at all to be honest, indeed I don’t see where you’re even contradicting any textbook description of central bank as lender of last resort.

    Couple of things: the Fed does not have to ALWAYS create base money in order to lend to other banks although even if it does that’s of little significance, secondly, other than in times of crisis (such as recently) this amount of lending is normally a very small proportion of total interbank lending.

    Hence, while the fed is forced to engage in continual adjustment of the base in order to defend its target rate (non controversial claim, everyone agrees with this), the Fed can still have major DISCRETIONARY changes on the base on top of this, either by changing this target rate, engaging in open market operations to influence other rates, or engaging in open market operations on the long end of the yield curve (quantitative easing).

    But again, I still don’t see who this is contradicting, I never made any particular claims with regards to the amount of discretion the Fed might have over the base. It has nothing to do with whether banks create money in excess of liabilities and capital (borrowing from the Fed is a liability, or it could be capital if money is attained from the Fed through other means), or merely through double entry accounting (which is non controversial, nothing new, not how economists define creating money and is what everyone agrees with).

  • Tom Brown

    My argument boils down to these essentials: The “money multiplier” is a myth because:

    1) The Fed publicly sets and then does what it takes to defend an overnight intra-bank reserve interest rate.

    2) We have a fiat money system with a floating exchange rate and the Fed can create (or destroy) fiat base money (MB) at will electronically, either to loan out to banks as reserves or to exchange (as reserves) for other bank assets. Note: it can also meet the demands for currency (M0, or paper bills and coins), but that is included in MB, and represents a tiny fraction of MB.

    The Fed uses 2) to accomplish 1). Since this is the case, reserve requirements could be discarded as they do not provide any meaningful limit on the expansion of the money supply in the broad sense (M2), since ultimately the Fed always stands ready to create and loan out reserves as needed to meet it’s interest rate goal. In other words, we no longer truly have a “fractional reserve” banking system, though we still have reserve requirements.

    Alternatively, the Fed COULD choose to try to control M2 through fractional reserve requirements and control of the base money supply. If it did this, it could not control the overnight interest rate and probably the payment clearing system would not run as smoothly… and there’d probably be a danger of bank runs. Plus, it may find that it does not truly have control over “the money supply” in the broad sense (M2, M3, M4, etc) because reserve requirements are only on household demand (checking) deposits and exclude business and time (savings) accounts, and the reserve fractions seem to be up for negotiation depending on the size or political influence of individual banks. But even assuming it were to adopt hard consistent reserve requirements, looking at the definition of M3 and M4… It’s not clear the Fed could control those.

    In summary, the Fed cannot control both the overnight interest rate and the base money supply. It has to pick one, and it has chosen (probably wisely) the former.

  • Jason

    Furthermore, in your system it’s still the Fed increasing base money, whether or not they have little choice in the matter, not private banks.

  • Greg

    Think about this Jason

    You run a bank. It just opened yesterday. You have $50,000 in deposits, you have $1,000,000 in equity (100 guys put up 10,000$ each for stock).

    Bill Gates walks in and wants to start a project in your town. He wants to borrow 100 million. He is putting up 20 million of his own and you determine that this venture will generate over 5 million a month in cash flow. You also find that his net worth to debt ratio is well over 1 billion. Does he get the loan? Of course he does. The only question you have about making the loan is the condition of the guy wanting to pay it back not how much you have in deposits or equity.

  • Cullen Roche


    You started your criticism by claiming there is a connection between the base and the amount of lending that is done. Now you seem to be saying the opposite….

  • The Undergrad

    Cullen, why is this such a contentious point among neoclassical economists? I’ve brought this issues up with many of my professors and they vehemently deny my claims, even after taking the time to read the literature I forward them. Can’t tell you how many professors I’ve shown your paper to but yet nothing seems to stick, or click for that matter…

  • Jason

    Which post are you referring to?

  • Cullen Roche

    One thing people hate is change. And when you confront them with a view of the world that forces them to totally change their views they reject it. I rejected all of this when I first confronted it. Hell, I rejected parts of MMT and MCT years after I had accepted it. I think it takes a real evolution of thought to come to grips with this stuff….It’s not easy.

  • Jason

    It’s probably because they define things differently to you. If when you say “create money” you mean double entry accounting, then those professors don’t disagree with you, they just don’t actually call it creating money. If you tell them that banks can extend money BEYOND both their liabilities (including central bank lending) AND their capital then they would disagree with you, only because what you’re saying is completely untrue.

  • Cullen Roche

    Your first post here today said banks create many multiples the amount of base money.

    “banks create many money that is many multiples of the monetary base, by definition”

    Then you said:

    “the Fed does not have to ALWAYS create base money in order to lend to other banks although even if it does that’s of little significance”

    The reality is that a banking system does not need much, if any, base money to operate smoothly. See Canada’s system where the base money supply is practically insignificant.

  • Jason

    That doesn’t imply that increasing the monetary base will make them increase their lending more so, it’s just stating things how they currently ARE, i.e. that banks are lending money which is many multiples the amount of base money in the system.

    Note, this is NOT the same as saying there is a linear or constant or direct causative relationship between base money and broad money, few economists actually believe this. This is NOT the same as saying that banks multiply the amount of base money always to a specific multiple implied by the limit of the money multiplier.

  • Jason

    You can’t survive without any monetary base. If a new country was created with the same laws but had absolutely no money whatsoever, a bank couldn’t just pop up and loan money into existence without any capital at all, there needs to be an initial amount of money first, society would probably create money or the government would start issuing some by fiat.

  • Cullen Roche

    So all these comments just to state that the banking system needs SOME level of base money in order for the Fed to achieve monetary policy and settle payments????

    I’ve never said anything otherwise….

  • Jason

    To clarify earlier, as I said there are two versions of the money multiplier, one that there is a constant relationship between base money and broad money (which few people believe), the other simply being that banks will lend an amount such that broad money will be a multiple of base money, but it wont necessarily be a constant multiple, and there wont be a linear causative relationship.

  • Cullen Roche

    That’s not really true. A banking system can exist without a reserve system or what we currently call base money. The USA has, at times during its existence, had no reserve system or national bank. And as my initial article said, these systems aren’t very stable. Still, your implication that the amount of base money constrains lending in some way, is totally wrong. Base money exists to settle payments, achieve monetary policy and meet reserve requirements (which don’t exist in all banking systems). It doesn’t constrain lending.

  • Cullen Roche

    Your latter assertion is basically a monetarists way of backpedaling out of the money multiplier while avoiding having to admit that the money multiplier model in all of their textbooks is wrong….If that makes a neoclassical economist sleep better at night then whatever floats your boat. Now we can all say “I win” (but really this is just another thing the neoclassicals have lost). :-)

  • Jason

    If you hold that banks are capital constrained, then if base money is sufficiently low, it constrains lending (unless every single bank does not keep any money as reserve… which can only be done if the central bank is continually providing more base money), and even then it still requires a tiny bit of base money to start off with (otherwise in what sense are banks capital constrained). Note, I’m talking specifically with the laws we have now in this hypothetical country, before recently there was plentiful base money without the need for a central bank: gold.

  • Tom Brown

    Jason, the reference you provided seemed to emphasize that in a fractional reserve banking system the process of banks making loans does NOT increase the base money supply (and that was their argument about why banks don’t create “money”… they don’t create BASE money). I’m pointing out that the supply of base money is a dependent variable (of the overnight rate and the demand for credit) in OUR system (rather than the fractional reserve system they discuss). You say you agree with this. And yes, I agree the amount the Fed is forced to create to defend its overnight rate is small… and I suppose you could claim that’s a function of the “reserve ratio” at least in the example I gave. There’s another example though in which the reserve ratio would not reduce the amount needed: the case of a depositor taking their funds out in the form of cash. But in that case I’ll grant you, the total demand for cash is also small.

    Now what you also seem to be saying is that the overnight rate is ALSO a dependent variable. I can see this argument if you mean that what the Fed does is actually target an inflation rate (the ACTUAL independent variable), or a combination of the inflation rate and the unemployment rate. But you seem to be saying that it could be targeting the base money supply here too. Or perhaps a broader definition of money (M2 perhaps?). No I checked… you said “base” money (that, I don’t get).

    The time lag is starting to get long here though, and the feedback loop less clear. An inflation rate/unemployment rate/money supply level is selected, and an overnight rate set to accomplish this. Months pass. The overnight rate is adjusted. etc. Is that what you’re getting at? (I’m ignoring QE here).

    The thing is that the Fed has really good control (through the base money supply) over the overnight rate. It can hit that target very precisely and the mechanism is clear. Expanding the feedback loop to include other meta-variables may be valid, but the picture gets a lot more murky, and the time lags start to get long.

    Ultimately it sounds like it’s a matter of semantics for you. Whether you call it money or not, banks can create (w/offsetting loans) deposits (denominated in US dollars) that can be used just like money in every sense whether you want to call it that or not. And ONLY banks have a license to do that. No other business. You seem to be saying “Well, yes, nobody disputes that, but the Fed ultimately has influence over how much through the interest rate they target, which they control through the supply of base money.”
    If that’s what you’re arguing… “influence” I won’t dispute that. But to restrict the definition of money to ONLY base money seems bizarre to me.

    Take a look at Scott Fullwiler’s article. He talks about Canada too. He claims in that article that most banks in Canada desire and are able to close out each day WITHOUT any reserve balance in their account overnight (I know you say above that they CHOOSE to have reserve balances). I don’t know which is true, but try writing to Scott to get his reference for that. I’ve had luck getting responses from him. In either case, the required reserve fraction there is zero, and thus the “money multiplier” is infinity! … yet they don’t have hyperinflation. So what in your opinion could possibly be going on? I guess if the overnight reserve balances are in general 0 (base money supply = 0), then it’s a case of 0*infinity for the broad money supply (ha!).

    I think Steve Keen has shown mathematically that a PURELY endogenous “money” system could function. In his first generation of models of the economy he excluded the central bank and reserves (he’s since added them to the model). I don’t have any problem seeing how this could function in the real world. If the government granted banks a charter to issue dollar denominated debt in the form of deposits (which could be used to pay taxes with), why not? I believe in his model an initial loan-asset/deposit-liability was created (out of thin air) on the bank’s balance sheet, the principal of which is never to be paid back by the firm that borrowed it, but which resulted in a fixed interest every year paid to the bank. I’ve studied the balance sheet style inputs he used for his differential equations governing the model. I didn’t see any problem. Is this model ultra simple? Absolutely… but it proved a point to me: “money” or whatever you want to call it can be created out of thin air (with offsetting debt) and used as a medium of exchange to run a simple economy w/o any “base money” whatsoever (except in the form of a definition). The deposit half of the balance sheet can be transferred multiple times and functions just like currency. It can be invested in assets which gain or lose value. The debt half (loan) is tied to a particular borrower and its value doesn’t chance and it thus has a different nature.

    That’s why I think it boils down to semantics for you. Whatever you want to call what banks create, they can create it, and it can be used just like money in the economy. Can the Fed influence this process? Absolutely… that I agree with.

  • Jason

    Why do you keep talking about reserves? I didn’t mention reserves there. Base money and reserves are not entirely the same thing.

    I don’t feel like you’re addressing my points either, please answer the question:

    In this new country, can a bank make a loan, even if it has NO CAPITAL and no deposits (there can’t be any deposits because nobody has any money to deposit yet)? If it can make this loan, it is not capital constrained, which would contradict what you’re saying earlier.

  • Cullen Roche

    What is the purpose of making up mythical worlds? We exist in a world where there are capital requirements for a reason. If banks don’t meet the capital requirements the FDIC knocks on their door on Friday night, asks for their books and the keys to the front door. So your example is totally irrelevant.

  • Cowpoke

    Let’s add some “Credit Card” tender to the fire…:

    Table Two: Total Debt Outstanding and Securitized Pools as proportion of GDP and Personal Consumption Expenditure:
    Current Dollars
    % of GDP
    % of Personal Consumption Expenditure

    The steady increase of unsecured debt is having real effects on the overall growth and trajectory of US economic expansion. The largest jump in unsecured debts contribution to the PCE, the its equivalent value relative to nominal GDP was from 2000-2001, rising nearly 2% in both cases.
    This shows the scale of private credit creation in the credit card market, as receivables moved off-balance sheet through ABS allows the lending pool to be re-capitalized using the same capital reserves (either through deposits for
    banks or initial capital stock used by specialist or non-bank lenders). Taken together we can see a plausible link between the widespread private credit creation used by consumer credit lenders and its impact on broader macroeconomic indicators via individuals’ use of debt to fuel consumption.”

    Translating macroeconomic trajectories of rising household unsecured debt levels to the dynamics of meso-level trends in the consumer credit industry is a complex and difficult process. Therefore, the remainder of this presentation will focus on the credit card industry and the particular dynamics of financialized competitiveness among card issuers.
    The US credit card market has a huge number of participants. Currently, over 6,000 depository institutions issue VISA and MasterCard credit cards and independently set the terms and conditions on their plans. Close to 10,000 other institutions act as agents for card-issuing institutions. In addition to the firms issuing cards through the VISA and MasterCard networks, two large nonbank firms, American Express Co. and Discover Financial Services, issue independent general purpose credit cards to the public. Yet, despite this highly competitive market profit rates remain extremely high. The Federal Reserves yearly congressional report on credit card banks (December 2006, seventeen banks with assets exceeding $200 million met the definition of a credit card bank) reported net earnings before taxes of 3.34 percent of outstanding balances adjusted for credit card-backed securitization. For example, for all commercial banks, the average return on all assets, before taxes and extraordinary items, was 2.01 percent in 2006.
    Crotty (2008) referred to puzzling coexistence of intense competition and historically high profit rates as ‘Volcker’s Paradox’. While Crotty was looking specifically at commercial banking, this presentation extends his framework to the credit card industry to examine what factors contribute to this phenomenon which seems to defy basic economic principles. In the case of the credit card industries two key components have contributed to high profit rates in the face of growing competition: the widespread use of ABS to create new credit without recourse to new capital or deposits and the unique pricing techniques facilitated through credit scoring and profit profiling.
    Firstly, securitization lowers the cost of funds for credit issuers, which increases profitability, and allows credit to be recycled through re-capitalisation of loan pools. Creating an ABS involves bundling together thousands of small loans, specifically the anticipated interest payments owed on these outstanding loans, into a Master Trust. This trust is a legal entity, called a special purchase vehicle, backed by a certificate which investors buy shares and receive interest and principal payments over a twelve-month contract. Through this process interest-based income streams are bundled together, sold, and transferred off balance sheet as a sale of assets. By moving assets off balance sheet the loan pool can be re-capitalized. The resulting re-capitalisation recycles the credit pool, allowing new loans to be issued from the same capital stock. Profits result from the difference between
    the high rates of interest charged on credit cards and the lower rate of interest paid out to securities.–The%20organizational%20dynamics%20of%20private%20credit%20creation.pdf

    This is a decent read..

    I think this sort of study shows that Credit markets created by banks using private sector participants shows how much they really create money.

  • Jason

    “But you seem to be saying that it could be targeting the base money supply here too.”

    You’re looking at it backwards, they can increase or decrease the monetary base, how they do this is dependent on what their policy is. I don’t maintain that it can ‘target’ the base money supply whilst being consistent with an interest rate target or an inflation target. However, if they were purely only targeting interest rates, then this would only require small adjustments to the base, in the meantime this gives them headroom to increase the monetary base by other means should they want to, without it impacting the interest rate. It may however prevent them from decreasing the monetary base too much.

    ” So what in your opinion could possibly be going on?”

    Remember, I DON’T agree with the form of the money multiplier that says banks will lend all the money they have until all that’s left is reserve (and hence, an infinite amount if they don’t need reserves, of course in reality they would always need some reserves, unless the central bank is continuously willing to lend more base money to banks to meet withdrawals and transfers, in which case its ultimately the central banks fault for allowing this infinite money growth).

    If in Keen’s model a bank can loan money into existence without having any capital, then it’s not capital constrained. Do you have any evidence that this would be legally possible? I’m not saying it couldn’t happen in a hypothetical economy, or some ancient economy.

  • Jason

    The existence of this hypothetical country should be perfectly possible under today’s laws if what you say about there not being any need for any base money whatsoever is true, it doesn’t matter if its unrealistic, its a necessary implication of that claim.

  • Jason

    Err, actually I’m not sure about my claim with respect to interest rates there, also I’m talking purely the overnight rate. Of course it would influence other rates.

  • Jason

    My bad, I did mention reserves, my point still stands however.

  • Cullen Roche

    I said if there was one national bank there would be no need for reserves. Obviously, in today’s system banks are required to have capital (of which reserves have to make up a component of the asset side of the balance sheet) and base money needs to exist due to the way the system is designed. I’ve explained this 100 different times.

  • Jason

    Also it seems you’re confusing reserve REQUIREMENTS with reserves. Banks don’t need reserve requirements, but they either need reserves, or ready access to money from the FED (HENCE WHY I QUALIFIED MY STATEMENT WITH “unless every single bank does not keep any money as reserve which can only be done if the central bank is continually providing more base money”, otherwise there will be bank runs.

  • Jason

    “I said if there was one national bank there would be no need for reserves.”

    Does physical cash exist in this world? Also, my main point is that there NEEDS to be base money under the current legal system (hence me saying a new country created with the same laws) in order to operate, even if its a tiny amount. My argument about banks wanting to keep reserves is secondary and not as important, base money =/= reserves.

  • Cullen Roche

    Jason, you keep comparing a modern monetary system to arcane gold based systems. You do realize that cash transactions in many countries are virtually non-existent, right? The world is fast moving to a system where bank customers removing money from banks is a pointless service. Sweden is practically already there. In 20 years there will be monetary systems that completely disprove your idea of “bank runs” and cash as having some superiority over a bank deposit. Bank runs are for monetary system from the old ages. They hardly happen any more and when they do happen it’s because the banking system is being poorly managed by the govt. You can keep reaching into the past for evidence that proves your point, but the reality is that we’re moving into the future and your monetarist based thinking is getting left in the dust.

  • Cowpoke

    Understand that there is a whole new section to the banking industry that is UNSECURED/CREDIT Cards.
    UNSECURED!!!!???????????? OH wait!! can we “Securitize them?LMAO!!

  • Geoff

    Excellent post, Mr. Roche. The idea that banks create money virtually out of thin air is hugely under-appreciated. But what do you make of Mr. Mosler’s view that banks are really part of the govt, i.e., they are federally regulated agents of congress?

  • Jason

    Except that ‘neoclassical’ economists have been arguing about this for decades, for instance Kydland & Prescott’s ” 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 aregenerally 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 evenmore than M2, with the lead being about three quarters. (p. 12)”

    They regard this conclusion as neoclassical; that’s because neoclassical economics is not defined by beliefs regarding institutional make-up or specific behaviour of banks, but rather a specific style of modelling (marginalist), which is why most modern ‘neoclassical’ models don’t even have the monetary base in it as a level for the monetary authorities, only the interest rate.

  • Cullen Roche

    That’s one of my major disagreements with MMT. I think banks are private for profit entities that operate mainly for private purpose. MMT says they are agents of govt that operate primarily for public purpose which makes the money supply state controlled in essence. I say the money supply is privatized in what is really the outsourcing of money control from the govt to private companies. It’s not state controlled at all. It’s a market based money system controlled by an oligopoly of private entities and supported by the govt.

  • Jason

    “Jason, you keep comparing a modern monetary system to arcane gold based systems.”

    And you keep attacking stawman arguments I never made.

    “You do realize that cash transactions in many countries are virtually non-existent, right?”

    If you’re talking about cash as in ‘hard cash’, I wasn’t specific to that (even though physical cash withdrawals and usage are still very common in every single country I have ever been in, in my entire life, including much of Europe). When I say withdrawals or transfers, I don’t necessarily mean in the sense of HARD currency, it could simply be an electronic transfer to another bank.

    “The world is fast moving to a system where bank customers removing money from banks is a pointless service. Sweden is practically already there. In 20 years there will be monetary systems that completely disprove your idea of “bank runs” and cash as having some superiority over a bank deposit.”

    What on earth are you talking about? I never once said anything about ‘cash having superiority over a bank deposit’.

    “Bank runs are for monetary system from the old ages. They hardly happen any more and when they do happen it’s because the banking system is being poorly managed by the govt. You can keep reaching into the past for evidence that proves your point, but the reality is that we’re moving into the future and your monetarist based thinking is getting left in the dust.”

    Bank runs don’t need physical cash, they can happen by people electronically needing to transfer their money to another bank or savings institution in excess of what the bank can provide. I never said bank runs were common place anyway, I explicitly said earlier that the idea of a bank run today is extremely unlikely. They may even be impossible, but that’s only because the Fed makes sure it’s impossible. I’m also not a monetarist.

    None of my arguments are contingent on physical cash.

  • Jason

    Or maybe you think reserves are physical bank notes? That is also wrong, or at least I don’t define reserves that way.

  • Cullen Roche

    Jason, I am trying to give you the benefit of the doubt, but your argument keeps changing or is incredibly unclear.

    Bank runs occur when a bank is insolvent. Not because it doesn’t have enough reserves. Who really cares if a bank is insolvent and you move your money to another bank? That doesn’t prove anything regarding your point. The amount of reserves or base money in the monetary system would remain the same. AGAIN, we’re back to a capital constraint.

    If you don’t think bank runs (of any kind) are prominent or even important then why do you keep falling back on this as “proof” of your point? This is just circular reasoning….

    Thus far, the only thing I can conclude from your comments is that you think base money is necessary in the modern monetary system. That’s a point I don’t disagree with. But it certainly doesn’t mean banks are reserve constrained or constrained in their lending by the level of base money supplied by the central bank….So I don’t really see where you’re going with all this….

  • Cullen Roche

    No, had you read the original post you would have noticed a clear distinction between “notes, coins and reserves”. Again, you don’t seem to have read the material, but insist on arguing your original point despite this….That’s fine, but I don’t really see the point of continuing if you’re not even going to read my post clearly or the cited material you explicitly requested in the first place.

  • Pierce Inverarity

    How is it not money?

  • mike


    Let me give you a real life example since I work on the lending side. I worked for a mortgage wholesale lender. We have a capital requirement of x million of dollars before a warehouse lender allow us to access a credit line in 10 to 20 X multiple of the capital we maintained. At no time do we actually use this capital to lend out since doing so would actually decrease your ability to lend out. Instead we lend (fund) loans with the warehouse lines and turn around and sell this loan back on the secondary market for a spread. In your opinion did we just create money? Was any money multiplier involved? Depository institution such as major banks have access to the cheapest source of capital around, mainly deposit. The deposit give them a capital base just as our verify capital gives us a capital base to work against.

  • Jason

    “Bank runs occur when a bank is insolvent. Not because it doesn’t have enough reserves. Who really cares if a bank is insolvent and you move your money
    to another bank?”

    This is incoherent. A bank being insolvent by definition means it’s unable to pay its debts. If it can’t service it’s liabilities it means it can’t borrow money or doesn’t have enough reserves or assets easily and quickly convertible into money to do so. Given this definition of insolvent (is there another definition? If so I’d like to know), I don’t see how the rest of your paragraph follows at all.

    “That doesn’t prove anything regarding your point. The amount of reserves or base money in the monetary system would remain the same. AGAIN, we’re back to a capital constraint.

    If you don’t think bank runs (of any kind) are prominent or even important then why do you keep falling back on this as “proof” of your point? This is just circular reasoning….”

    What point or proof are you referring to? Please quote specifically the argument you are talking about. It honestly seems like you’re arguing against somebody else, I don’t see what this is in regards to at all.

    “Thus far, the only thing I can conclude from your comments is that you think base money is necessary in the modern monetary system. That’s a point I don’t disagree with. But it certainly doesn’t mean banks are reserve constrained or constrained in their lending by the level of base money supplied by the central bank….So I don’t really see where you’re going with all this….”

    I don’t really know what ‘reserve constrained’ means even though I hear it touted all the time, it seems to be used entirely inconsistently. My only point is that a bank needs money (capital) before it can lend money. If there is absolutely no money initially, a bank can’t loan money into existence under the current system. I’m also reading the FED paper right now and it doesn’t so far disagree with this at all, it makes no mention of banks creating money out of thin air, it just refers to other sources of funding than normal deposits. Where specifically do you think I’m saying banks are ‘reserve constrained’ (I never used that phrase once), again please show the specific sentence.

  • Cullen Roche

    You win. Banks need capital (some of which are reserves, ie assets, which means they are capital constrained as my original post said).

  • Jason

    I know you don’t mean that but I guess this debate has gone on far too long and hasn’t got anywhere, we’re just talking past each other so now would be a good time to just end it. It’s been fun, ciao

  • Cullen Roche

    No, I really do mean it. We both agree that banks need capital to operate. I just think we’re disagreeing because of a semantic point which relates to the fact that reserves are an asset for banks so I get your point that banks need reserves, but I think neoclassicals overstate the importance between the link in lending and base money.

  • Jason

    Oh, well that’s good I guess. The main reason I feel so strongly about this stuff is because we have this group called “Positive Money” , and they are adamant that banks create money out of thin air without needing any capital (well, not all of them, but much of their followers at least) and of course I think they are dead wrong. They often link to here or MMT as a source when I engage with debates with them on forums, so I keep imagining you guys as agreeing with them on this, but I guess I’m wrong on that.

  • Pierce Inverarity

    Your strong feelings also come across as really prickish too over the internet. You’ll find an engaging audience here if you engage us in a condescending manner.

  • Pierce Inverarity

    *if you DON’T engage us in a condescending manner.

  • AndyCFC

    Look up the history of the Commonwealth Bank of Australia that wasnt capital constrained as was set up by the state.

  • Dave

    Some thoughts on money creation and MR:

    The distinction between inside and outside money (in the sense of MR) is too narrowly considered. Not only banks, govt and the fed can create money. All companies that manage in some way money (insurance companies, funds etc.) that are neither part of the fed nor the govt nor the banking system can create money. Every company can and does create money by simply giving credits to their suppliers and customers and every individual can create money by giving credit to another individual without any bank, govt or fed involved. So every process where credit gets granted (buy today, pay another day) creates money. And so does every settlement, payment, bankruptcy or maturity “destroy” this money.
    So this distinction is more or less interesting for accounting purposes but it doesn’t say or explain much about credit and money creation.

    One should better distinguish between money the fed can or can’t control directly (the term of inside and outside money would still be appropriate). And one should distinguish between “inactive” or “active” money. Money that is either hoarded, saved (inactive) or money/credit that is productive in circulation (active; to buy goods and services) which has an inflationary character. These matters are much more relevant to monetary theory than the mere issuer of money. And it is more relevant to determine whether the money/debt creation is inflationary or neutral and then to distinguish between the different types of inflation.
    A discussion about if the fed is a private or govt institution seems to me obsolete since it has the character and elements of both entities.

    Then an important fact is ignored. Liquidity. Liquidity is not defined by the amount of money in a system but by “credit/money and its maturity”. We all know the case when “sound” companies with many assets go bankrupt because the money is not available at maturity.

    These are just a few thoughts on a very complex theme. Unfortunately many people think that the problems in understanding the monetary theory are just a matter of accounting. But we have empirical prove that accounting is by far not as reliable as it should be and that one can easily enhance balance sheets.

  • Geoff

    Agreed. In fact, many people would probably argue the opposite, that the Govt is an agent of the banks :)

    Seriously, I think MMT does deserve a ton of credit for hammering home these two points:

    1) Banks aren’t reserve constrained
    2) Loans create deposits

    If you grasp those two points, you understand how banks create money out of thin air. The light goes on.

  • Dave

    Geoff, money from banks is never created out of thin air since every credit granted is backed by the goods and services you buy with this credit… And when you pay back your credit or default it will be settled and the money supply decreases again to the previous amount before your credit was granted.
    The problem is if you default or your backed good (i. e. home) purchased with credit loses value the credit has either to be written off or down. If this doesn’t happen, as experienced in the current crisis, the overall liquidity decreases. But it should be clear that if you default there will be capital (workforce, commodities etc.) become available since the goods are not (fully) backed anymore.

    The fed’s famous printing press, if monetizing govt debt, is another matter though…

  • Johnny Evers

    This is probably a dumb question, but this is a fascinating conversation.
    How do we define ‘reserves’? For example, when a bank makes a loan it must go out and find reserves — are these deposits at other banks? Or does the Fed create the reserve? And if a bank needs reserves, even if it’s for a foolish loan, must the Fed provide the reserve?

  • Dave

    The reserves are from deposits and physically held currency. If you pay $1’000.- in your deposit the bank will have to increase reserves. If you draw money the bank can decrease reserves.
    If the bank doesn’t has enough primary liquidity they will have to lend at another bank (fed fund rate). If this is not possible they will lend the money from the fed at the discount rate. If this is not possible they will draw on secondary liquidity (sell securities).

    As you can imagine the problem with the reserve requirement are if Bank A gives you a credit of $100’000.- and transfers it to another Bank B the reserve requirements drop at Bank A but increase at Bank B. So the bank which gives you credit actually benefits from taking that risk and Bank B is punished because reserve requirements increase.

  • LVG

    It looks like most of the confusion in this thread comes from the fact that reserves are an asset for the bank and banks need assets to operate in today’s world. So reserves make up a portion of the capital base and since banks are capital constrained it appears like reserves make it possible to lend.

    Just remember, reserves are an asset. That’s a component of capital = assets – liabilities.

  • LVG

    Reserves are deposits held on “reserve”. So, when a bank takes in customer money they must hold some on “reserve” for various purposes. Reserves are deposits held at the Fed. Think of reserves as cash held in a special market for banks. They’re essentially loans from the Fed to the banks.

  • LVG

    MMT makes the same wrong argument that the other economists make. They say banks “leverage” state money as IOUs. This is like saying there is a money multiplier, but in a way that they can also reject the money multiplier. The language in MMT is a disaster. That’s a big reason why it’s not catching on.

  • Tom Brown

    Yes, but Bank A in your example, must transfer reserves to Bank B in the amount of the deposit. So it’s Bank A’s responsibility to come up with those reserves to transfer. After that, Bank A’s responsibilities are over, and you are correct, Bank B is responsible to maintain reserve requirements.

  • Tom Brown

    You wrote: “Every company can and does create money by simply giving credits to their suppliers and customers”

    However, that form of “money” isn’t transferable is it? Person A can take their bank created bank deposit (from a loan) and use it to buy something from person B who then spends it on persons C, D, and E, etc.

    Not every company can create that kind of credit… credit denominated in US dollars. I know I can’t! I can credit somebody by telling them I’ll do $X of work for them in the future, but they can’t take my IOU and use it to buy groceries!

  • Tom Brown

    Capital requirements are specified by CARs (Capital Adequacy Ratios). The CAR is the ratio of “capital” (Tier 1, Tier 2, etc) to risk weighted assets. Because reserves are weighted by 0 (no risk), you are partly correct… having reserves reduces the denominator and thus increases the CAR (the CAR must be above a certain threshold, typically about 6% to 10%). However, I don’t believe it’s absolutely necessary to have reserves to meet the CAR thresholds. Say a bank had a single mortgage on its balance sheets and nothing else. A mortgage is weighted by 0.5. Does the mortgage qualify as Tier 1 or 2? I don’t know, but if it does

  • Geoff

    Dave, we are talking about the ability of the lender here (the bank), not the borrower. And just because the money can disappear as easily as it is created does not negate the point.

  • Cullen Roche

    Yes, corporations who sell debt are raising capital. They are obtaining purchasing power from other people in exchange for securities (a claim on future income). This is not the same as creating purchasing power in the form of a loan. It is transferring purchasing power.

  • Bond Vigilante

    The distinction between “Outside money” and “Inside money”is artificial. It’s both credit. The only difference between the two is that money is credit without a maturity and zero interest.

  • Tom Brown

    You wrote: “Geoff, money from banks is never created out of thin air since every credit granted is backed by the goods and services you buy with this credit…”

    I disagree, but perhaps it’s just a matter of semantics. If you go into a bank and get a loan… say a bank with a blank balance sheet, the resulting balance sheet has a loan-asset and a deposit-liability of equal amounts (thus the sheet is still in balance). It’s *almost* like the bank created, out of thin air, +$ and -$. That description isn’t accurate, however, because the +$ are transferable and are essentially money in every sense. The -$ however, are not! They represent a loan to a particular individual (you) and they do not rise and fall with the value of assets purchased with the +$. They rise at a predefined rate of interest. They can disappear (as you point out) in the case of default, or in the case of the loan being repaid.

  • Cullen Roche

    No, the importance of the distinction is understanding where it originates.

  • Tom Brown

    Good point.

  • Tom Brown

    OK, points taken.

    Re: adjusting the monetary base: OK, I can see that… for example with QE, the monetary base is technically increased through asset swaps… but this swapping has little effect on the real economy (in times of low demand for credit from qualified borrowers). And this can be done w/o impacting the overnight rate too much, at least at the zero lower bound.

    The “monetary base” is a bit of an arbitrary definition though, don’t you think? What really matters is how much money circulates in the economy, not how much excess reserves sit in bank reserve accounts.

    Re: the infinite money multiplier… yes, I recalled that you understood that the money multiplier only defines an upper bound right after I hit “Add Comment.”

    S Re: Keen’s model, I’m sure the bank wouldn’t satisfy the Basel Accords (ha!) with the initial loan… *perhaps* after it collected enough interest payments, but certainly not initially.

    So, overall, I don’t think there’s much we disagree on except language perhaps.

  • Cowpoke

    Now I feel like I am back to square one after reading all this Plus this:

    Some points to note:
    “The general idea here should be clear: while an individual bank may be able to decrease the level
    of reserves it holds by lending to firms and/or households, the same is not true of the banking
    system as a whole. No matter how many times the funds are lent out by the banks, used for
    purchases, etc., total reserves in the banking system do not change. The quantity of reserves is
    determined almost entirely by the central bank’s actions, and in no way reflect the lending
    behavior of banks.”

    So banks do/can lend “EXCESS reserves” because they simply end back up as reserves on either thier account or another banks reserve account.

    “What about the money multiplier?
    The idea that banks will hold a large quantity of excess reserves conflicts with the traditional
    view of the money multiplier. According to this view, an increase in bank reserves should be
    “multiplied” into a much larger increase in the broad money supply as banks expand their
    deposits and lending activities. The expansion of deposits, in turn, should raise reserve
    requirements until there are little or no excess reserves in the banking system. This process has
    clearly not occurred following the increase in reserves depicted in Figure 1. Why has the money
    multiplier “failed” here?
    The textbook presentation of the money multiplier assumes that banks do not earn interest on
    their reserves. As described above, a bank holding excess reserves in such an environment will
    seek to lend out those reserves at any positive interest rate, and this additional lending will
    decrease the short-term interest rate. This lending also creates additional deposits in the banking
    system and thus leads to a small increase in reserve requirements, as described in the previous
    section. Because the increase in required reserves is small, however, the supply of excess
    reserves remains large. The process then repeats itself, with banks making more new loans and
    the short-term interest rate falling further.
    This multiplier process continues until one of two things happens. It could continue until there
    are no more excess reserves, that is, until the increase in lending and deposits has raised required
    reserves all the way up to the level of total reserves. In this case, the money multiplier is fully
    operational. However, the process will stop before this happens if the short-term interest rate
    reaches zero. When the market interest rate is zero, banks no longer face an opportunity cost of
    holding reserves and, hence, no longer have an incentive to lend out their excess reserves. At
    this point, the multiplier process halts.”

    So is this multiplier effect opposite in the way it is presented here?
    The Money is created first through loans and then the reserve reqirement rises AFTER the loan is created.
    So am I understanding this correctly, that Excess Reserves can be loaned out?

  • Bond Vigilante

    Even then is the distinction is artificial. Because the FED is a bank like any other bank. And it creates like any other bank “money/credit out of thin air”. It’s meant to be “a lender of last resort”. Yes, it swaps bonds for reserves. But the FED first has to create that money “out of thin air” (=print money) to be able to swap those bonds for reserves.

    Commercial banks don’t issue money, they issue credit. The FED is the only legitimate issuer of money/currency. Guess what would happen when every bank would issue its own money ? Then there would be several currencies inside the US.

    But the government (excluding the FED) is a currency user and therefore has to borrow the money like any other currency user. The government isn’t a bank and therefore can’t create credit out of thin air.

  • Tom Brown

    Those are good questions. It’ll be interesting to see the answers. However I think of it this way (which may be wrong!): only banks, and the Treasury have reserve accounts. Banks can loan their reserves to each other, but they don’t loan their reserves to any other entities. I’ll admit this is a bit of a semantic game, but I think it’s totally consistent to say that. When you or I borrow from a bank, the bank creates a loan/deposit. If we transfer a deposit to another bank, the originating bank must transfer a reserve amount matching the deposit to the other bank’s reserve account. In doing so, it may overdraft it’s own reserve account, and then be forced to borrow reserves on the money market (perhaps even from the bank receiving the deposit) or the Fed to cover the overdraft. So when banks loan each other reserves it’s a different process than when they make loans to other entities. They actually do transfer the reserves from one account to the other in the case of intra-bank reserve loans. The loan shows up as an asset on the lending bank’s balance sheet and a liability on the borrowing bank’s balance sheet (of course the borrowing bank’s reserve account increased to offset this — provided it wasn’t used to cover an overdraft!, and the lending bank’s reserve account likewise decreased). That’s my understanding!

    Now if you were to take out a loan, and then withdraw your deposit in paper currency the bank would give you vault cash, which is a form of reserves. So in that case, yes, you would be walking away with what used to be the bank’s reserves (once you have it in your hands, it’s no longer considered “reserves” but it’s still “base money”). It won’t re-enter the system as reserves again until someone deposits it back in a bank account. In the counter-intuitive language of accounting (as far as I know, I’m not an accountant!) you would say that the lending bank “credited” it’s reserve account, since reserves are a bank asset, and when assets go down, they’re “credited.” When they go up, they’re “debited.” The opposite terms are used for increases/decreases in liabilities. The bank may then want to purchase more vault cash from the Fed from its reserve account, but that’s just swapping one form of reserves for another. A “cash advance” from your credit card would count as a form of this kind of “vault cash” loan.

    BTW, that bit of accounting lingo (if I’m correct about that) made me think that if an accountant ever tells you he wants to fully credit your life asset… it means he’s going to kill you!

    I know you were stressing “excess reserves” but I didn’t see that in the original piece you quoted, and I’m not sure how “excess” modifies what I wrote above.

    One thing I’d like to know is what banks can do with their reserves other than hand them out as vault cash, transfer then from one form to another (vault cash to/from electronic reserves), loan them to other banks, process payments between bank patrons, and buy Treasuries with them as primary dealers from the Treasury (or buy Treasuries from the Fed? Does the Fed ever re-sell any of the Treasuries it buys up in QE for example?)

    Let’s go back to your “excess reserves” qualifier: Can banks use those excess reserves to buy treasuries on the open market? Can they use them to buy foreign government bonds? Can they use them to buy anything they want?

    The reason I’m interested is because I always hear Michael Hudson say things like “Banks have taken all those excess reserves that Bernanke has pushed on them [yes I understand they were swapped for other securities], and used them to buy Brazilian government bonds!” Is that actually possible? To attempt to answer my own question, I guess I’d say “why not?” I just view a reserve account as a bank account for a bank. Can’t those funds be used for anything, provided the bank continues to meet it’s reserve requirements?

  • JJTV

    When it comes to reserves and lending I always say: “There is no balance sheet transaction that can loan a Federal Reserve liability to a non-bank entity”. I have yet to find one at least.

  • Romeo Fayette

    I think this is the rub for a lot of people: the idea that banks are enabled to create money by law. They disburse a loan and counterbalance the loan asset with a deposit liability–ex nilho.

    When Cullen continually cites “loans create deposits,” I think a lot of people get confused, thinking ‘Oh, when a bank disburses a loan, the recipient bank account has a deposit.’

  • Tom Brown

    “Loan” you say, but how about just transfer, like for a purchase? Say Bank A wants to use some of its capital (in the form of reserves) to buy donuts for a meeting. The donut shop’s deposit account is increased. If the donut shop’s deposit account is at Bank B, then Bank A’s reserves are decreased and Bank B’s reserves are increased in the process, correct?

    Thus can’t that process be used for almost anything? For example, to speculate on derivatives, commodities, or foreign debt?

  • Romeo Fayette

    So, a bank is enabled, by law, to create money. When a bank makes a loan, that new loan asset is counterbalanced by a deposit liability, created ex nilho (essentially).

    In its dying days, why didn’t Lehman set up a bunch of SPVs off balance sheet, lend them money (which loan asset creates an offsetting deposit on LEH’s balance sheet), have the SPVs turn the loan money around and inject equity back into LEH to give it Tier 1 capital? Legal reasons I suspect?

  • jt26

    I think if the banking system is involved then this may not be true. E.g. buying corp debt on margin or repo, or bank origination (e.g. like MBS origination).

  • jt26

    A question closely related to reserves is, why do you need reserves (CB reserves) at all (pros/cons)? Some countries, e.g. Canada does not have reserve requirements.

  • Tom Brown

    I’m anticipating that the answer might be “to meet capital requirements” (i.e. Basel Accords). Although Scott Fullwiler strongly implied in his article “Krugman’s Flashing Neon Sign” that in Canada, they actually don’t have reserves, like you state (overnight anyway):
    “Consider, for instance Canada, which has no reserve requirements and where the central bank is so good at forecasting banks’ demand for reserve balances (due to how the interbank market functions there) that banks actually desire to hold no reserve balances overnight—reserve balances only exist on an intraday basis.”

  • Tom Brown

    Lehman did do some funny business near the end… look up “Repo 105″ or “Repo 108.” These were techniques to make it’s balance sheet smell better temporarily.

  • SS

    Lehman did do some of this. But in the end, it’s just shuffling the deck chairs on a sinking ship. Loans don’t create net financial assets. So there’s no capital creation in this process. It’s zero sum in essence. So if Lehman lends itself money then the bank is taking the other side of the loan that it ultimately sends back to itself. Loans aren’t capital because they create an asset and a liability.

  • Cowpoke

    Thanks Tom,
    This paper discusses “Why are banks holding so many excess reserves?”

    So the whole premise of them even asking the question makes me question do they (Banks) have to hold reserves.

    In the description of “reserves” in this paper, here is how they describe them:

    “1 Reserves (sometimes called bank reserves) are funds held by depository institutions that can be used to meet the
    institution’s legal reserve requirement. [B]These funds are held either as balances on deposit at the Federal Reserve or
    as cash in the bank’s vault or ATMs.[/B] Reserves that are applied toward an institution’s legal requirement are called
    required, while any additional reserves are called excess.”

    based on my interpretation, banks have a required reserve amount. However, once the FED started paying interest on reserves banks started parking EXCESS reserves for the slam dunk guaranteed interest pmt by the FED. So I take it that Excess reserves can be loaned out, just not Required reserves.

  • Romeo Fayette

    Play with me here so I get this…
    LEH could give its SPV a perpetual term loan, and as I said, the SPV turns around and injects equity into LEH. That loan doesn’t have to be paid back “to itself,” you know what I mean?

    I’m using this as a real world example, but I guess what I’m saying is: can’t LEH just create inside money to address all obligations?

  • SS

    Banks have to hold reserves because the Fed makes them hold reserves. And also because reserves are how the Fed implements monetary policy. But don’t confuse banks holding reserves as meaning that they lend their reserves.

  • SS

    Let’s say LEH sets up a SPV and then lends a bunch of money to it which it then injects back into LEH. You haven’t created any net financial assets in this transactions. You’ve just passed the hot potato. Lehman has the asset in the loan and the liability of the new deposits. The SPV has the assets in the deposits and the liability of the loan. No net financial assets. Lehman wouldn’t be getting anything in net because there’s no capital created from these transactions.

    There’s no point in lending yourself money.

  • SS

    The point I am making is that Lehman’s balance sheet doesn’t improve simply by making loans. It improves when it receives the loans back in full PLUS interest. That’s the primary way a bank increases its capital. Or by raising money from outside entities.

  • Cowpoke

    They don’t lend “Required” reserves, but they can lend “Excess” reserves.

  • Tom Brown

    Yes, I guess it wouldn’t make sense to loan reserves if you were just meeting the reserve requirements… then you’d just have to borrow them from someone else. So excess reserves must be loanable to other banks at the Federal Funds rate. Still, that doesn’t mean they loan those excess reserves to any entity except another bank.

    JJTV’s statement above is not inconsistent with that.

    Back in the days when the Fed wasn’t paying interest on reserves, and the Federal Funds rate wasn’t zero… then it was in a bank’s interest to attract transfer deposits, since those funds could be used to meet reserve requirements at a lower interest rate than what banks would have to pay each other (i.e. the Fed Funds rate was more than the interest paid to depositors). New deposits would free up reserves for lending to other banks at the higher overnight rate.

    SS (below): I think you have to make an exception for intra-bank loans, otherwise the Fed has to do all the Federal Fund rate loaning, and I don’t think that happens.

  • Tom Brown

    … to another bank.

  • Romeo Fayette

    Yea, that’s where I’m confused though (I might be messing up my balance sheet accounting here): LEH has a loan asset offset by a deposit liability PLUS new equity. SPV has a loan liability and an equity stake in LEH. Since LEH has the legal power to create money, it’s created the money it needs I mean, can’t it just skip the SPV altogether, create a deposit and use the offsetting “money” asset to pay its obligations?

    I guess what I’m asking is how do you create a net financial asset? The SPV wasn’t capitalized, but neither are some shadow banks. Further, in my mind, if LEH were able to attain capital from Warren Buffet (for example), that’s a net financial asset to LEH, but it just endogenous money already within the system.

  • jt26

    Good point about the capital. Perhaps the the Canadian system is more insular (less volatile; the Big 5 are very dominant) and has higher capital requirements, so they are already well-reserved. In some sense, the international USD banking system (Eurodollars), who have no Fed reserve requirements, also seem to function, so the need for official reserves is not very obvious (each bank has to reserve). Maybe the reserves are historical, when interbank lending wasn’t as smooth as it is now ;-)

  • Tom Brown

    But every bank doesn’t HAVE to issue it’s own currency. Why would they do that when they have a charter to issue legitimate US money (equivalent) in the form of dollar denominated deposits offset by loans on their balance sheets. Their charter also allows them to make good on those deposits (should they be transferred, used to purchase goods, or withdrawn in the form of paper currency) by borrowing dollars from the Fed funds window if need be (or on the intra-bank money market at the same Fed Funds rate).

    Thus their business model is supported by the difference between the interest rate they charge and the interest rate they pay for those funds.

  • Tom Brown

    When you write “LEH has a loan asset offset by a deposit liability PLUS new equity.”

    The “equity” shows up as a liability, correct?

    Before “capital injection”:

    Loan to SPV
    SPV deposit

    deposit with LEH
    loan from LEH

    After “capital injection:”

    SPV loan
    SPV equity

    Equity in LEH
    Loan from LEH

    The “deposit” entries disappeared from both balance sheets because it was essentially transferred to equity.

  • Tom Brown

    In the “after capital injection” case for LEH, you might argue that you need two more entries:

    Reserve balance from SPV to purchase equity
    Loan of reserves from Fed or other bank to cover SPV’s withdrawal of deposit to purchase equity

    … but those two cancel each other out rapidly when the reserves are transferred back to pay off the loan. Since this is loan of reserves by a bank, I believe that’s the way it would appear on the balance sheet (in this weird self-reflective case).

  • Tom Brown

    I guess it’d be more clear if I’d written “Borrowing of reserves…” rather than “Loan of reserves…”

  • Cowpoke

    Tom, you are a lot more knowledgeable than me on this but why do I keep reading that they CAN loan them out?
    See here as well from the Cleveland FED:

    “While banks cannot control the overall level of excess reserves, there are a several ways they can reduce the level of excess reserves on their own individual balance sheets. [B]They can lend excess reserves to other banks in the federal funds market, they can lend them to consumers or businesses, or they can purchase securities.[/B] Each of these outlets has been constrained for various reasons since the recession.”

  • Amin

    Banks do create money, but only money backed by central bank currency.

    This means that the amount of credit that a bank can issue that will be accepted by others (i.e. money) is limited by the market’s perception of their ability to obtain central bank issued base money.

    The legal requirement to back private credit with government currency gives control over the long term money supply to the government. Private bank operations can create fluctuations in the money supply, but those fluctuations that kept by market forces within the bounds created by the government through its control of the monetary base and central bank lending operations.

  • Tom Brown

    That’s a great question! … and I don’t think I’m more knowledgeable than you. That’s a great question for Cullen. Here’s my theory, for what it’s worth:

    When you hear people here emphatically say “banks don’t lend reserves!” … I agree that’s a consistent world view, but also it may be a bit of a semantic game. … When the bank creates a loan/deposit it stays like that until that deposit is actually used for something. That can often be IMMEDIATELY! When you pay for something with a credit card, that’s the case. In those cases, if the payee happens to have an account at the same bank… it’s a cinch! No reserves needed: just debit one account and credit the other. But in the much more likely case that the recipient’s account is at another bank, then the lending bank has to come up with reserves to transfer to the other bank in short order. So if the bank has those reserves (excess reserves) no problem. If not, they’ve got to go borrow them.

    So when the author of the article you quoted talks about loaning out excess reserves “to consumers or businesses” perhaps the mechanism above is what he’s referring to. Practically, most of the time, a loan will result in the lending bank having to transfer reserves to another bank.

    So you can still state that “banks don’t lend reserves” … but they do go borrow them if they need them! I think the point is to get people to stop thinking that banks wait around for deposits so they can then go ahead and loan those out. In other words, that’s a way to emphasize that banks meet the demands for credit, and are not hamstrung by a lack of reserves. They still must meet capital requirements (and guess what… reserves count as low risk assets – a good thing – in those requirements!), but ultimately capital requirements have more to say about the quality of the loans on the bank’s balance sheet (I believe… look up Capital Asset Ratio or Capital Adequacy Ratio in wiki… they’ve got a pretty good summary). They are a balance sheet requirement. Reserve requirements are only requirements on *some* kinds of demand deposits.

  • Tom Brown

    Another theory is that the author you quoted is confused. ;^)

  • Tom Brown

    I mostly agree, however I really think you’re overstating the Fed’s ability to control the broad money supply. The Fed could TRY to do that, but then they’d give up control over the overnight interest rate (in times of high demand for credit). Now you might claim that the Fed is really trying to control something else, like the inflation rate (or the unemployment rate, or the rate of growth of the S&P 500 or the growth or housing prices) by adjusting the overnight interest rate… but that mechanism is a lot less clear, and the time lag can be months between updates of the overnight rate.

    In times of low demand for credit, the Fed can certainly increase the base money supply… but that money supply (aside from the small amount of physical currency) is not what circulates in the real economy. Here again, I think the Fed’s ability is severely limited. It can create lots of excess reserves, but it can’t force the banks to lend. If the population is desperately paying down debts to private banks (and thus destroying money in the process)… I believe the Fed’s attempt to increase the broad money supply by swapping bank assets for reserves won’t do much.

  • Cowpoke

    Thanks Tom for helping me through this.
    Some more FAQ:
    “Why does the Federal Reserve want to pay interest on excess balances?
    Paying interest on excess balances should help to establish a lower bound on the federal funds rate by lessening the incentive for institutions to trade balances in the market at rates much below the rate paid on excess balances. Paying interest on excess balances will permit the Federal Reserve to provide sufficient liquidity to support financial stability while implementing the monetary policy that is appropriate in light of the System’s macroeconomic objectives of maximum employment and price stability. For more information about the implementation of monetary policy with the payment of interest on required reserve balances and on excess balances, please see the”

    This part”Paying interest on excess balances should help to establish a lower bound on the federal funds rate by lessening the incentive for institutions to trade balances in the market at rates much below the rate paid on excess balances.”
    I have seen multiple times and do not understand.
    I think this is the same investment speak about the Yield and price of treasuries. I have always gotten confused on that as the Yield Falls the price goes up or Vis Versa somthing like that.

  • Cowpoke
  • Cowpoke

    Aight, I think I have some more clarification:

    “The demand for Federal Reserve balances has three components: required reserve balances, contractual clearing balances, and excess reserve balances.”

    Tom, I apologize for being abit anal on this and you are correct that semantics play a large part and I think that has been the biggest and continues to be the biggest issue in our Nation/Planet.
    As I asked last month in the question section about Words/names. Remember Shakespeare’s “Rose by any other” name….
    I think the second tier education system using a differing vernacular to name things differently but are still the same thing causes a lot of confusion.

    Thanks Tom for working with me through this convoluted process of learning.

  • Cowpoke

    I feel like a Dumb Arse! I just remembered that My In Laws have a close friend that Works At The FED.
    Hell I am going to ask him to get me the answer..

  • Romeo Fayette

    Good choice to put things in Fullweiler’s balance sheet terms. While the equity shows up as a liability, it’s also cash on the asset side, which it looks like you indicate with your “Reserve balance from SPV to purchase equity.” (Just to be clear.)

    So in this case, I think what you’re suggesting is that the Fed would have LEH shut down because it couldn’t find inside money to cover its overdraft borrowings that the Fed provided it intraday to settle payment, right? But, I don’t think LEH would have that problem since noone in the interbank market will know its distress since it’s not capital constrained–it’s raised the capital (and can repeat the raise with infinite SPVs) subject to covering the Fed’s overdraft with an interbank loan.

    As SS points out in his reply below, the problem is: why is this a viable business model? There needs to be net interest margin in the mix to justify the charade. But, you don’t think Dick Fuld would’ve done this to keep his doors open one more day, one more week, one more month, until market dislocations (in his mind) normalized? I know all about Repo 108… these guys would’ve done anything.

  • Romeo Fayette

    This is a healthy conversation btw. It’s good to vet these ideas. Thanks

  • Pierce Inverarity

    Also, consider, LEH didn’t have a legal ability to create money. It wasn’t a chartered bank. It was a relic of the pre-Glass-Steagall Act, a true investment bank that couldn’t afford the lawyers that came up with the genius idea that Morgan Stanley and Goldman took to turn themselves into bank holding companies in order to 1) access the Fed window and 2) potentially “create money”.

  • Pierce Inverarity

    Gotta agree with Cowpoke. Thanks for your thoughtful articulation of banking from the government to the regional level for all of us. You’re very skilled at articulating these mechanics in ways the rest of us can understand.

  • Romeo Fayette

    I guess this brings me to my point… While LEH has covered its obligations via these shenanigans, they have not created money.

    Since we haven’t settled the LEH example yet, just consider any bank that creates money in the way described by Cullen or Fullwiler. In these examples, the bank has an asset offset by a liability, and ultimately, the only form of net creation is derived from exogenous “borrowings from the Fed,” which is an overdraft that ultimately has to be covered by the interbank market (preexisting/endogenous cash). Thus, the exogenous creation is ultimately withdrawn from the system when the overdraft is paid off with interbank borrowings. That where I think the Fullwiler argument (and maybe Cullen’s) falls apart: he assumes Fed overdrafts are covered in the money market. First, that’s under the assumption that the money market isn’t frozen and hasn’t locked your bank out. Second, it’s ultimately drawing upon endogenous money. Fullwiler seems to suggest that a bank’s net interest margin on a loan is further money creation, but again, that money (NIM) is derived from preexisting, endogenous money.

    As far as I understand it, the only means of true, permanent creation is via interest on Treasury securities (Tbills/notes/bonds) and IOER… I don’t think IOER should be considered though, because the Fed Funds rate is higher & therefore withdraws reserves on an aggregate basis. I guess I should also include the creation derived from a Fed that’s willing to pay above FMV to buy MBS/Govies fm a PD as part of QE. I guess there are fiscal measures as well–tax policy can affect sequential levels. How else does the Fed input outside money and increase the system’s reserves?

  • Cullen Roche

    The private sector is a closed system. It can’t create net financial assets. So even though a bank can create money it can’t create net financial assets. Only the govt can do this.

  • Amin

    Good points about the Fed’s constraints given its focus on interest rates, but I’d argue that this is a matter of intentions rather than control. The Fed, even when attempting to keep overnight lending rates
    within a certain band, is still exerting influence over the money supply. It’s simply not attempting to meet a particular money supply objective while doing so.

    That the money supply is a secondary concern to the Fed doesn’t change the fact that it’s the Fed’s decisions that ultimately determine the
    available of the monetary base that backs all bank credit.

    Regarding base money not circulating in the real economy: that’s certainly true, and is the reason the Fed cannot control fluctuations in the money supply.

    I think importance of these fluctuations is eclipsed by the importance of the long-run average money supply though, and I think the long-run average money supply will trend toward an equilibrium multiple of the monetary base, which market fundamentals (e.g. risk of default), rather than the decision of any bank, determines.

  • Dave

    As some already mentioned before the banks are not reserve constraint. The reserves are only to back the deposits but not the credits…

  • Dave

    …so Bank A does not necessarily have to transfer reserves to Bank B. There are many ways to disburse this $100’000.- for a bank. If the bank really doesn’t have $100’000.- it can sell securities so it changes a low rate security to a higher rare credit which the bank will favor.

  • Dave

    Dear Tom, the credit given by a company is given to purchase goods and services (usually the goods and services of that company). If you need bread but you don’t have enough money you won’t go to Bakery A to ask for credit but then you purchase your bread at Bakery B… The main thing here is that these credits from companies are used like any other credit/money to buy goods and services (buy today and pay another day) which can have inflationary character.
    The company granting credit to buy their goods will have to spend the money to produce those goods (where payment will be settled in the future) on other persons (suppliers A, B, C etc.) as well…
    A home loan from a bank is to buy a home and not to give a friend who wants to buy a Ferrari but still your money will end up at different persons…

    And don’t forget, companies can issue corporate bonds which is credit as well.

  • LVG

    Corporate bonds are one way a company raises capital. It does not create new purchasing power. It exchanges it from other people. Selling bonds is a cheaper form of financing than taking out a loan. The company gets to set the terms rather than going to a bank who sets the terms. But these securities are not the same as borrowing from a bank. A bank can actually issue new purchasing power. A corporation will generally take it from someone else.

  • Dave

    BTW you wrote:”Not every company can create that kind of credit… credit denominated in US dollars.”

    The $ is just the arithmetic unit used to calculate the value of a good like you measure in foot or meter… 100 years ago you used gold to measure the value of a good but it was still denominated in $.
    So a company can express its credits in $, yen, € or whatever is appropriate to them… Most likely a company in the US granting credit to a US customer will express the value of the credit in $. So yes, every company can create that kind of credit denominated in US$.

    Hope that helps…

  • LVG

    That’s not true. Corporate securities like stocks and bonds require someone else to give up cash in order to obtain the securities. A loan doesn’t require anyone to give up anything. The bank doesn’t give up cash from its vaults and the borrower gets deposits. This is nothing like a corporation issuing debt or equity.

  • Dave

    LVG you have to consider that corporate bonds are probably bought with savings (inactive money) but the company will use the money raised with the bonds to finance a project. So the money will be activated and it will by in circulation. So it can create purchasing power.
    If a bank uses excess reserves to buy bonds it will activate money which will be in circulation and create purchasing power.

    As I mentioned before it is rather more important to define if money has inflationary character (active or inactive) than to distinguish between the issuer.

  • LVG

    “inactive money”? You’re just making stuff up now.

  • Dave

    LVG, it is not necessarily to give up cash to buy stocks or bonds. There are several forms to finance anything…
    As mentioned before one can use savings (inactive money) which then will be used to finance a project. Money goes suddenly into circulation.

    Person A selling a stock to Person B might just cash in to buy a car. This looks like a swap of money from Person A to Person B but not if Person B uses a credit to finance the purchase…

    As written before a bank can use excess reserves to buy bonds…

  • LVG

    Of course the purpose of selling securities is to obtain cash. That’s the WHOLE purpose. When a company goes public by selling stock they are raising money in exchange for cash. So, you buy the Facebook IPO, Facebook gets cash, you get a stock certificate that represents a claim on future cash flows.

    You’re misunderstanding the basics of the way a cash market works.

  • LVG

    raising money in exchange for stock. Sorry.

    And yes, when you buy a bond you give someone cash and you get a bond. That’s how the market works!

  • Dave

    You do not seem to understand the difference between money that is inactive, unproductive and money that is active, productive and the relation between the two…
    You can sell your home for $100’000.- to buy FB stocks. Most likely the person buying your home will take out a loan.
    The bank granting the $100’000.- loan creates deposit.
    So you give the $100’000.- to FB which invests (purchases goods) this money and creates therefore purchasing power. The money supply rises $100’000.- until the loan gets settled.

    You have $100’000.- savings (or in a vault as gold). You then decide to buy FB stocks with that money. The money gets activated which means the money is suddenly used to purchase goods and services which before it did certainly not…

    Cash payments are by far not the only way to buy stocks and bonds.

  • Romeo Fayette

    Cullen, SS & Pierce:
    Completely agree that the private sector doesn’t create net financial assets. However, I do not understand how it creates “money”? When a loan is disbursed, the bank’s liability goes from a deposit (belonging to the borrower) to some form of capital (whether interbank borrowing or debt issuance). Even if that deposit is funded/supplanted temporarily by a Fed overdraft line, a source of private sector funding must cover the overdraft by COB. Unless I misunderstand your definition of “money,” there’s no creation of money there, just the recycling of preexisting private sector cash (i.e. an interbank borrowing that must be continually rolled).

    Perhaps you’re suggesting that a bank can make a loan, which makes a deposit on its balance sheet; then, were it to end there–without the borrower or the borrower’s payee withdrawing the loan disbursement from the originating bank–the bank only needs to make sure it has/finds ample reserves against this deposit. Say that’s only a 10% reserve requirement, which means the bank has created 90% of that in new money. Is that what you mean?

    So it’s because of fractional reserve banking that banks can create money? Even if the borrower in the previous example withdrew the loan from Citi to buy a house, and the house’s seller takes payment and deposits it in BoA, I guess you’re suggesting that Citi (the originating bank) will just end up borrowing these BoA deposits overnight to fund its loan… one big circuitous loop, over and over, a tangled web of such arrangements spun throughout the system? (After all, if BoA has already made a loan which that deposit subsequently funds, then the deposit must exist somewhere else in the system, and some bank will have it idle & available in the money market to cover Citi’s funding need. Right?)

  • SS

    See new thread below.

  • SS


    There is a difference between purchasing power like bank deposits and net financial assets. Banks create purchasing power. Technically, anyone can create purchasing power, but your purchasing power isn’t accepted everywhere. For instance, you can create Romeo Notes and start distributing them. But no one will accept them. So they’re not very valuable as a form of purchasing power. What is valuable as a form of purchasing power is bank deposits. Bank deposits are US dollar denominated purchasing power. It has value because the government says it has value and because the private sector has created a huge amount of output that backs these deposits.

    But purchasing power is not net financial assets. When a bank creates a loan it creates deposits which create new US Dollar denominated purchasing power. But it does not result in an increase in the private sector’s net financial assets.

  • Cullen Roche

    Right. Banks are outsourced the power (by the govt) to create new USD purchasing power. The money supply is privatized in the USA.

  • Dave

    Either the credit is backed by the goods you purchase or if you just have a credit for fun you’ll have to pay it off anyway and the credit is backed by your own workforce… It doesn’t matter how gave you the credit…
    You forget that the lender will check your solvency, creditworthiness, guaranties and so forth to be sure the credit is even backed (partially) if you default.

    A bank credit is always backed since a credit produces demand. Otherwise you wouldn’t need a credit.

  • Dave

    Tom, as I wrote before you will need only a credit if you want to purchase goods or services you can’t afford right now. If you don’t want to purchase anything you don’t need a credit. It is as simple as that. You seem to forget why we have a credit system…
    So every credit will be backed by the goods and services you purchase with it… If you have a credit to buy a car it is backed by the car… If you have a home loan the credit is backed by the home you purchase… If you get a credit to buy a loaf of bread your credit is backed by that loaf…
    Otherwise why would you get a credit for lets say $10’000.- if you won’t purchase anything worth at least $10’000.-? You wouldn’t need a credit if you wouldn’t buy anything. There is no bank credit out of thin air. Even if you get a credit just for fun for $10’000.- you will have to work off that credit so it is at least backed by your own workforce…

    Hope that helps to understand my point… But not to mix up with the fed monetizing govt debt…

  • Dave

    Other than MR-theory I distinguish between “active”, “inactive” and “neutral” money. So money that creates demand (active) or money that is just savings (inactive) or money that has the character of a “swap” which I call “neutral”.
    This is the key to understand inflation or deflation… Active money is the money in circulation and creates demand for goods and services, inactive money doesn’t create demand. A bullion for example in your vault is inactive since it doesn’t produce any demand for goods and services (as long as it is in your vault) so it can’t have inflationary character…
    A credit usually has inflationary character since it creates demand for goods and services…

    MR theory distinguishes between the money issuer which doesn’t tell you anything about the usage of that money. It doesn’t tell you if this money creates aggregate demand or not. So it doesn’t tell you if it has inflationary character or if it is neutral on aggregate demand.

    If I made that up fine… so I just invented it… but if you think logically it makes much more sense then to distinguish between inside and outside money (the issuer of money)…

  • Romeo Fayette

    Ok, I see what you’re saying here. I guess I didn’t fundamentally disagree with you at all, because we’re both saying that “purchasing power” has been created–but it’s a liability tied to an asset–hence it’s not a permanant (in a loose sense) net new financial asset.

    Can you address a few of the points in my last reply though please?

    First, the bank still needs to fund this new “purchasing power” (inside money), and that funding is drawn from somewhere else within the inside system, like the interbank market. So, it’s effectively swapped some other bank’s purchasing power for the purchasing power it’s created itself. Zero sum, right?

    Second, because of fractional reserves against deposits, I can see how money is created when a customer takes a bank loan and maintains his deposit at the lending bank. Am I correct here?

    Finally, if a customer takes a bank loan and a deposit is made at another bank (not the lending bank), no purchasing power has been created because of the aforementioned zero sum reality. Is this correct?


  • Tom Brown

    Sorry I haven’t been on this thing all day!

    Thanks for the complements Cowpoke and Pierce… but again I emphasize that I’m REALLY no expert. I’m just parroting back to you what I learned from Cullen, Keen, and Fullwiler… in addition to my own ruminations on the subject I guess. Cowpoke, please let us know what your friend at the Fed says!

  • Tom Brown

    Dave, that’s not how I understand it. Of course I get my understanding from others, such as Scott Fullwiler here (check out figure 2 and the description):

    So Bank A does transfer reserves to Bank B in this case in the amount of the deposit… that doesn’t mean its “reserve constrained” because it can always borrow those reserves to cover the overdraft (bank A overdrafts its Fed account in Scott’s example when it makes the transfer). It could borrow those reserves anywhere… even from bank B that it just made the transfer to (B of course would need to continue to meet its reserve requirements).

    So I think I’m being consistent with Scott when I say that A needs to transfer reserves to B. If that’s not the way it works, then Scott (and I) are both wrong!

  • Tom Brown

    see my comment (and Scott’s article above, especially the “borrowings” in figure 2 to cover the transfer in a case where Bank A doesn’t have it).

  • Tom Brown

    I especially like that last paragraph.

  • Romeo Fayette

    I completely get that and I agree. But, the spread comes from a debtor paying back interest, which is paid back with preexisting money from within the system. Nothing has been created here. Further, as I’ve said/asked in my last two replies, Fullwiler’s balance sheet (and intuition) says that a loan must be funded in the interbank market if the borrower transfers the loan proceeds from the originating bank to another. Because of this, I don’t see how money is created UNLESS THE LOAN PROCEEDS REMAIN AT THE ORIGINATING BANK.

    (Not yelling, just emphasizing.)

  • Cullen Roche

    Don’t think about it in the micro. Loans, in the aggregate don’t really get paid back. The interest to pay back loans really comes from new loans. It has a ponzi aspect to it though that term isn’t entirely appropriate because it implies unsustainability. The banking system, as we’ve constructed it, is quite sustainable (assuming there is economic growth and value to sustain the value of all those loans!).

    Payments settle in the interbank market because there is a multi-bank system. If there was one national bank you wouldn’t have this interbank classification. So I think the existence of this market is confusing you. Think of it like this. The reserve system exists to help independent banks pool funds so they can coordinate payment settlement. Let’s say you take out a loan at JPM and buy a new car. The auto dealer then deposits the check from the loan at BAC. How does BAC settle this payment if they don’t have the funds on hand? They borrow from the Fed or from BAC. The interbank market makes it so that the deposit at BAC is EXACTLY like depositing it at JPM. It integrates the banking system. So, it’s a lot like having one big nationlized banking system except we’ve managed to achieve this streamlined payment settlement process while maintaining private competitive banking.

  • SS

    New loans means banks must find reserves. These reserves either come from the central bank or from existing reserves. Required reserves will always be expanding as new loans will require new amounts of funds for settlement. So the central bank has to make new reserves available if they don’t already exist in the banking system. The act of lending forces the central bank to make reserves available in order to maintain the payment system and ensure that banks can settle all their payments.

  • Jason H

    Exactly… creating money/debt to fund increasing production, technology, research, efficiency, etc is a good thing (increases material wealth, increasing supply of goods & services, saves time, which offsets & decreases inflation)…

    creating money to speculate on commodities, destroying things, gambling,derivatives, etc is a bad thing since that decreases supply and/or increases prices

    P.S. To the other Jason, banks can & do ‘create’ money. Reserves don’t limit lending(money creation) since banks can get whatever reserves they need from the Federal Reserve after-the-fact… ask any banker..

    and the Federal Reserve has the power to create as much reserves as needed into perpetuity forever (in it’s charter as well as video Congressional testimony by Fed chairmen to ignorant Congressman)

    also, the Federal Reserve charter has mandates to create “full employment, production, and price stability” [aka low inflation]” (it’s failed at #1, #2, but ok on #3)
    it’s board of governors is appointed every 4 yrs by the President & it’s charter specifically states that if the Federal Reserve & gov disagrees, the Federal Reserve must defer to & is subservient to the Secretary of the Treasury, who himself is replaceable at anytime by the President

    The problem is that the Treasury Secretary is now Timothy Geithner, who is a lackey for Goldman Sachs & was a stooge of W. Bush’s Paulson(Goldman Sachs executive) & Clinton’s Rubin (Golman Sachs executive also), who all pushed for the deregulation on mortgage securities, CDS, & commodities speculation as well as the bailouts. For comparison, the Savings & Loans fiasco of 1980s went into FDIC receivership while Sweden’s banking system was nationalized

    –BTW, China banned derivatives & avoided the 2008 finacial crisis (Singapore & Japan only started to allow them in 2007 so their problem was small & nipped)

  • Tom Brown

    Yes, that’s true, however, the amount that must be created is scaled down by the reserve ratio, and only for those deposits which require reserves. For example, if a new loan creates a $1000 deposit, and there are no excess reserves available anywhere else to borrow (I’m ignoring capital requirements for simplicity), then the Fed only needs to loan (to the bank) $100 of new base money to cover the bank’s reserve requirements (assuming a 10% reserve ratio requirement).

  • Jason H

    Having been educated/trained as biochemist/molecular biology at UCBerkeley, there’s a physical biological reason for this, similar to “old habits die hard” & “you can’t teach an old dog new tricks”.. biologically speaking, “old beliefs die even harder than old habits” & why it’s important what’s first taught in schools (in this case, economics classes)

    When antiseptic procedures were first introduced circa 1847 by Austrian gov doctor Dr. Semmelweiss with his studies showing how death rates dropped by 90% at his gov university hospital by requiring doctors to wash their hands in antiseptic,
    the other medical establishment doctors ATTACKED him & rejected his antiseptic procedures
    it meant that their previous advice & practices of not washing their hands & not being antiseptic was actually killing & harming most of their patients (just like the deficit hawks/current economists advice is harming the economy now) –it would mean the ‘docs’ being wrong & harmful –and no expert likes to admit they are wrong or professionally harmful/incompetent

    It took about 40 years later after decades of fighting before the idea & practice of antiseptics became accepted

    Similarly, when hi-protein diets were first introduced for fatloss by Dr. Atkins in the 1970s & Dr. Atgatson (South Beach Diet) & studies showing how they caused TWICE the fatloss of hi-carb/low fat diets, most “authorities’ in the medical & nutritional community attacked them .. it took about 30 years before hi-protein diets were accepted as superior for fatloss & just as healthy hearth health as low-fat diets because most “experts” hate to admit that what they’ve been preaching of ‘low-fat’ diets for the past 50 years has been wrong & inferior

    It also used to be taught that “a calorie is a calorie is a calorie”, which is absolutely false because new Princeton Medical center studies show that after 6 months, subjects fed a hi-fructose corn syrup diet WITH THE SAME CALORIES as a control group gained 48% MORE FAT even though calories were the same

    This is a far cry from the corn-association funded studies that lasted only a few days or a few hours(!) that showed no difference between hi-fructose corn syrup groups & control groups

    –similar studies funded by cigarette companies decades ago that lasted only a few days to a few weeks showed no difference in cancer rates between smokers & non-smokers

    I mention this because it used to be argued for DECADES that cigarettes don’t increase cancer rates
    and it’s still being argued among some ‘experts’ that “a calorie is a calorie is a calorie” when studies show that hi-protein diets cause more fatloss & that hi-fructose causes more fatgain, even when calories are kept the same.

    It takes 30-50+ years to change myths because it takes that long for old dinosaur “experts’ to DIE OFF OR RETIRE so that their beliefs no longer dominate policy or are taught

    Every thought/belief by the brain is a neural connection synapse pattern … and the more often you think it & repeat it, the stronger the neural connection(s) gets…

    it’s like muscle memory, it’s also ‘belief’ memory..

    it’s especially strong when it’s first formed (this is why your first love, your first religion, your first political beliefs, etc is strong & very memorable)

    Anything that conflicts with previously held beliefs is usually rejected & disbelieved due to a “loyalty” to your prior & current beliefs

    studies show it’s takes about 21 days of repetition to form strong neural connections (21 days of repetitions to form a habit or muscle memory or beliefs, which is why boot camp & religious indoctrination retreats usually last 3 to 4 weeks miniwum)

  • Tom Brown

    Cullen, I think you have at least one error in your example. You write: “How does BAC settle this payment if they don’t have the funds on hand? They borrow from the Fed or from BAC.”

    That doesn’t make sense. One of the “BACs” should be a “JPM.” In particular, I think it’s the 1st one that should be changed into “JPM.”

  • Cullen Roche

    Yes, I mixed that up. Sorry. Should say:

    “Let’s say you take out a loan at JPM and buy a new car. JPM issues a loan which will create a deposit. The auto dealer then deposits the check from the loan at BAC. How does BAC settle this payment if they don’t have the funds on hand? They borrow from the Fed or from JPM. The interbank market makes it so that the deposit at BAC is EXACTLY like depositing it at JPM. “

  • Tom Brown

    Cullen, like I said I like the paragraph, but there’s one point that I’d clarify perhaps…. in the case of a TRANSFER deposit, what’s transferred is both an asset and a liability: reserves in the asset column, and the (cheap) deposit in the liability column. So if a bank needs reserves to cover reserve requirements or Fed overdrafts, attracting transfer deposits is a cheap way to obtain those reserves (cheaper than a loan from the Fed or another bank).

  • Tom Brown

    Are you sure? I think it should read:

    “How does JPM settle this payment if they don’t have the funds on hand? They borrow from the Fed or from BAC (or some other bank… or they attract transfer deposits from other banks). The interbank market makes it so that the deposit at BAC is EXACTLY like depositing it at JPM.”

    JPM, is the one that needs to “settle the payment” because the funds are coming FROM JPM to the car dealer with the account at BAC. I believe this is completely consistent with Scott’s “Figure 2″ in the “Krugman’s Flashing Neon Sign” article. It’s Bank A that needs to borrow the reserves… not Bank B! What he doesn’t point out is that Bank A could borrow the funds from Bank B to cover the overdraft (to make a nice tidy closed system for his example).

  • Cullen Roche

    Maybe I am not being very clear or having a brain fart. Let me elaborate in a bit more detail so this is super clear.

    If you take out a loan from JPM and pay the car dealer by check then the check made out to the dealer is deposited with BAC. So BAC would accept the deposit (deposit out of thin air). JPM needs to clear the check still. After all, that’s what the loan is, right? A promise to clear payments that you can’t make without them. The check will go through clearing and settlement on BAC’s side and JPM will use its reserves to clear the payment. Should it not have sufficient reserves it can find excess reserves on the overnight inter-bank lending market (Federal Funds Market in the US) – and don’t forget the deposit receiving bank now has excess deposits from accepting the check. And if all else fails they can use existing loans/treasuries as collateral and get reserves from the FED.

    Does that make more sense? On second read, it looks like I got the settlement wrong before….

  • Cullen Roche

    Yes, you’re right. I had settlement transfer backwards. Sorry. Two glasses of wine and everything goes to shit! :-)

  • peter fairley

    Prof Hanke of J Hopkins U. says Bernanke is only creating ‘state money’ and the more important ‘bank money’ has fallen 8%. 09/27/12 Bloomberg podcast following his publication.

  • Romeo Fayette

    This is the point if origin I’m trying to get to. How does the CB create reserves? Through what mechanism? As I understand it, the Fed shuffles excess reserves from banks that don’t need them (for minimum reserve requirements) to banks that do via the Fed Funds mechanism.

  • Romeo Fayette

    …and funds provided by the CB to cover overdrafts must be repaid intraday (before COB), which banks accomplish through interbank market. Where is the money creation here? the net new financial assets?

  • Tom Brown

    To illustrate my point, just look at Figure 2 in Scott Fullwiler’s “Krugman’s Flashing Neon Sign” article. We can easily modify that example: Assume that the person is Person 1. After Bank A makes a loan to Person 1, it must obtain reserves to meet its reserve requirements. If no reserves are available (either to borrow or through deposit transfers) from other banks, then Bank A must turn to the Fed discount window to borrow reserves equal to 10% of the deposit. Now say Person 1 spends the entire deposit on a car he purchases from Person 2 (Cullen’s example below) by writing a check. Person 2 deposits the check into his account at Bank B. Bank A only has 10% of the reserves on hand to transfer, so it runs a 90% reserve overdraft with the Fed. After the check clears. Say the original loan (and price of the car) was $10k. Then after the check clears Bank A has a $9k overdraft at the Fed, and Bank B has $10k of reserves. TEMPORARILY the Fed had to create (by loaning out more base money in the form of reserves) another $9k to cover the overdraft, for a total of $10k in reserve base money created by the Fed. However, Bank A must clear the overdraft ASAP to avoid monetary penalties. It’s also in Bank B’s interest to loan out its excess reserves. Bank B has $9k of excess reserves… a perfect match for the loan that Bank A has to repay to the Fed for the overdraft! Thus its in Bank A and Bank B’s interest for Bank A to borrow back $9k in reserves from Bank B to repay the Fed with… so lets assume that happens. (of course Bank A could borrow reserves from the discount window to cover the overdraft as well — but again, according to Scott, not without a penalty). Bank A still has the original loan for $10k on its balance sheet to Person 1, and presumably it can make money off the “spread” between this loan, and the $1000 of borrowings it has from the Fed to meet the reserve requirements in the first place (when it 1st made the loan), and the $9k of reserve borrowings from Bank B to repay the overdraft. Bank B, on the other hand, has a $10k deposit to Person 2, a $1K deposit with the Fed to meet its reserve requirements, and a $9k loan of reserves to Bank A. Its less clear if Bank B is making money from a spread. It certainly is on $9k (since it charges Bank A more interest than it pays to Person B on that $9k), but it well may be loosing money on the spread of that final $1k, since it must pay Person B, but the Fed may or may not pay it interest on its required reserve holdings. Does anybody know if the Fed currently pays interest on REQUIRED reserves (not excess reserves). Overall, since 90% of it’s “assets” are making money there’s a good chance that overall it’s making money, or close to breaking even anyway. The REAL spread is still with Bank A since it could charge a premium interest rate on the car loan, which makes sense because it took the risk of originating the car loan in the 1st place.

    So, overall, $10k was created in offsetting loan deposits in the “real economy.” Of this the Fed had to 1st create $1k, then another $9k (temporarily) to cover the overdraft, for a total of $10k created by the Fed, but then after the overdraft was paid back, only a net of $1k was created by the Fed. So $10k is created by the bank to purchase goods in the real economy, and this induces a net increase of $1k in base money that the Fed must create.

    In this example I’ve completely ignored capital requirements for simplicity.

  • Cullen Roche

    The Fed makes reserves as loans to the banking system as needed. They are powerless in choice here as they must supply any needed reserves to the banking system in order to keep the payments system functioning. So the banks lead the way here. It is inside money that determines the amount of necessary outside money. Not the other way around.

  • Tom Brown

    See my example below. Essentially they can be borrowed right back from the Bank that received the transfer. So the overdraft represents only a temporary increase in base money from the Fed… except for the 10% of required reserves. The Fed does increase (permanently) the base money supply by this much.

  • Cullen Roche
  • Tom Brown

    The world is whole again! Thanks for reviewing. BTW, in response to SS and Romeo above, I expanded your car example in excruciating detail. I hope its all correct!

  • Tom Brown

    Also, after writing to Scott he clarified that it is possible for the overdraft to be left on the books for more than one day, but more penalties are charged.

  • Jason H

    As Cullen/MR points out, the federal gov is a conditional user of the currency but the gov can always fund it’s spending by using primary dealers as well as the Federal Reserve as it’s own bank.. the Federal Reserve is NOT independent of the federal government & is actually subservient to the federal government & has no choice to but to fund & issue/create credit(money) to fund the federal government’s spending

    The Federal Reserve’s board of governors is appointed every 4 yrs by the President
    it’s charter specifically states that if the Federal Reserve & gov disagrees, the Federal Reserve must defer to & is subservient to the Secretary of the Treasury, who himself is replaceable at anytime by the President

    Also, the Federal Reserve has the power to create as much reserves as needed into perpetuity forever (in it’s charter as well as video Congressional testimony by Fed chairmen to ignorant Congressman)

    also, the Federal Reserve charter has mandates to create “full employment, production, and price stability” [aka low inflation]” (it’s failed at #1, #2, but ok on #3)

  • Cullen Roche


    The only thing I’d say differently is that the Fed is (primarily) designed to serve the banks, not the govt. Our market based monetary system is designed around the private competitive banking system as opposed to being designed around the govt. Important distinction in my opinion.


  • Tom Brown

    But just to be sure, the banks and the rest of the financial services industry make lots of campaign contributions (perhaps “crowding out” contributions from other sectors) and spend lots of money on lobbyists…. and as the % of profits generated in our economy shifts more and more towards the financial services sector and away from industrial capitalism this tendency will continue to grow. John Bogle, founder of the Vanguard family of mutual funds, has estimated that traditionally (in past decades when industrial capitalism was dominant) only 10% to 15% of profits went to financial services (which he stated is essentially overhead on the economy… much like the accounting department at an industrial firm such as Ford or Microsoft)… but today it’s over 40% (I recall him citing 42%). When asked by an audience member at one appearance if he’d recommend a career in financial services, he recommended they instead do something useful with their lives! ;^) … really! Funny statement from a guy who’s spent his life in the business.

  • Cullen Roche

    Right. But like I always say, I am not here to tell you why the system is the way it is. I am just here to tell you how the system is designed. In the USA, we have a system by the banks and for the banks. A lot of that is due to the fact that bankers have played a huge role in designing the system over the years….

  • Romeo Fayette

    Very valuable stuff here guys. Good convo, thanks.

    Thanks to Tom Brown’s comments above ( and his subsequent example, I think we almost have it hammered out…

    If Bank A originates the loan, the funding necessary to payoff the Fed overdraft can be borrowed on the interbank market from the recipient of the loan proceeds, which are in deposit form at Bank B. So, Bank B loans 90% (because of 10% RRR) of the capital to Bank A, who covers the overdraft. This raises 2 questions, both of which have been addressed but not sufficiently answered in this comment thread:

    1. From where does the other 10% come to payoff Bank A’s overdraft? The discount window is secured financing, which becomes insufficient as a bank’s loan assets are of lower quality and encumbered. The discount window accepts all kinds of collateral assets, but only investment grade quality (

    2. Tom Brown says the Fed permanently increases the monetary base to make reserves available for this 10% shortfall. I ask ‘through what mechanism,’ and Cullen says ‘the Fed achieves this via loans to the banks.’ Loans are not permanent. The reason “permanence” is important here is because the Fed doesn’t lend unsecured (with the exception of emergency programs), which means our Bank A is still struggling to find unencumbered assets to pledge at the discount window–as in #1 above.

    As you can see, there’s this problem of where the other 10% comes from to cover Bank A’s overdraft. As the bank’s lending loop pyramids up, Bank A won’t have unencumbered/quality assets enough to pledge for a discount window loan. Spread won’t cover this shortfall, neither will yields on Treasuries/IOER, which are the only source of Fed-injected outside money I can think of.

    So, answer the question of where the 10% comes from and/or by what mechanism the Fed increase the monetary base… then I think we’ll be done with this exercise.

  • Tom Brown

    Romeo, first off, great comment/question. I definitely won’t be able to answer definitively, however, I’ll make a couple of points:

    In my car example above that missing 10% ($1000) was a loan from the Fed discount window to cover the reserve requirements (RR) from the original loan. You point out that those loans need to be “secured.” Three things:

    1) In my correspondence (1 email) w/ Scott on this subject he pointed out that borrowing from the discount window to cover RR or overdrafts was “frowned on” and penalized, except for certain “seasonal adjustments.” He also stated that overdrafts CAN be held more than one day at the cost of additional penalties, and that collateral is needed for discount window loans, but that if it isn’t available, a heftier penalty is charged. So one potential path is the frown/penalties/”seasonal adjustment” path… i.e. it can happen, but there’s some pain/timing matters, that perhaps slow this process down a bit.

    2) Imagine that instead of the car loan above, we’re talking about a loan for an asset which is rising in value, like a house (during the bubble). Does this help with / lessen penalties for the “securitization” of the discount window loan? I don’t know.

    3) Perhaps this is where, again in boom times (when the bank’s stock is rising), the bank can offer equity or bonds in the bank itself as collateral. And since it’s stock is rising, this ability will tend to regenerate itself for a new loan down the road somewhere.

    OK, now I’ll sit back and await the real answers. Cullen? ;^)

  • Tom Brown

    Romeo, also, you write “loans are not permanent” … but isn’t it possible that an overnight loan can be taken on the money market every single day? (i.e. the loan keeps getting renewed on a daily basis). I believe that in this manner, many loans are never really payed back, including those taken from the Fed. I think Cullen has referred to this and called it a “Ponzi” mechanism (although he admits that the implication there is overly negative). Search for Ponzi here and you’ll find it.

    Also, another thought occurred to me regarding that 10% securitization: The bank will eventually make some money through interest payments. I already mentioned that the bank could put up equity or bonds in itself (based on the promise of those interest payments), but the payments themselves are very high quality (reserves). And who pays them? The people taking out the loans, so in my example, Person 1. Perhaps Person 1 uses the car as a cab to make money carting Person 2 around, and thus some of his deposits shifts to Person 1’s deposits (in my effort to keep my example self contained I’m starting to get silly here). What does this do? I guess it would just shift some of the “pain” of obtaining reserves to Bank B (to cover the transfers… while Bank B’s reserve requirements would actually shrink, but by only 10% of the transfer amount).

    Ultimately though, I see where you’re going with your question, and I think it’s mainly only a problem (explaining that missing 10%) when demand for credit is high (like during the expansion of the housing bubble). So I just can’t help but think that the rising asset prices (houses) in those times have something to do with “loosening up” some of the reserves from the Fed. Perhaps not a direct relation, but by through some mechanism.

    Also, if Person 1 or 2 have export businesses, they could be running a surplus and bringing funds into their accounts that way (when the funds are transferred, I’m imagining a similar mechanism w/ foreign trade: transfers involve a transfer of reserves-assets to back the deposit-liabilities).

  • Tom Brown

    Ah! Here it is… the bank repackages its loans as CDOs, gets the corrupt (or at least mal-motivated) ratings agency that it (and the other banks) hire to stamp it AAA (exactly equivalent to pre-2011 US gov debt), and voila! You’ve got “investment grade” securities for collateral… or to sell to Germans as an “export” from the good-ole USA! ;^)

  • Tom Brown

    In case anybody’s interested, I went back and looked at Fullwiler’s answer to my question about what determined whether a bank would choose to (i) borrow from another bank or (ii) borrow from the Fed discount window in order to cover an overdraft of its Fed reserve account (e.g. resulting from a deposit transfer or a payment settlement). Scott’s response:

    “(ii) is at a penalty unless collateralized. Even then, it’s at a penalty, just not as high. Also, under the current system, the Fed “frowns” on banks borrowing from it like this, and because of this other banks/investors/lenders view borrowing from the Fed as a sign of trouble (aside from seasonal borrowing and other special programs). In other words, since the Fed doesn’t like to lend to cover payment settlement like this, when it does do it others take it as a sign of weakness on the part of the borrower. To not send that signal to others or to the Fed (since the Fed’s a regulator, too, of course), they prefer not to borrow from the Fed.”

    So Romeo, there’s the “seasonal adjustments [borrowings]” I was referring to. I have no idea what those are, or what the “special programs” are!

    Here’s something else that might interest people as well: Scott had a few words about capital:

    “Assets are never capital. Capital is basically equity, so it’s on the opposite side of the balance sheet. (Some measures of capital–type II–include long-term bonds, too.) People get confused here because capital is sometimes referred to as “capital reserve.” Also, the “loan loss reserve” provision on the bank’s asset side confuses people–but that’s just the bank “reserving” part of its loans–marking them down essentially–to cover for anticiapted losses. It’s just an non-cash accounting entry sort of like depreciation. But reserve balances in the bank’s account at the Fed are very different.”

    By “type II” he may have meant “Tier II.” It took him about 4 months to answer my questions. I sent him a set of follow up questions… it’s been about 4 or 5 months now and I’m still waiting. Don’t get me wrong, I’m extremely pleased he answered my questions!… it’s just that he’s no Cullen Roche (with timely answers to a myriad of questions from the general public).

  • Tom Brown

    It appears that Steve Keen has tackled this subject as well in a very recent post… however, I’m not able to “Register for Free” for some reason at business insider, so I haven’t read it yet:

  • Cullen Roche
  • Tom Brown

    I see your point, and agree for the most part. However, it’s not quite as cut and dried as you make it. For example, I can get a “cash advance” from my credit card to add funds to my bank account. Perhaps I do that because I’ve got a lot of bills that automatically draw funds from my bank account, but I don’t know the precise amount… I just know that I’d like to have a “buffer” there (during a particular month or two when liquidity is tight) so I don’t overdraft my checking account and ruin my credit score, and I don’t have to go through the hassle of canceling the bill automation, and then re-starting it again with all the services (water, trash, etc.).

    Or here’s another possibility: Say I’ve decided to make some home improvements, and my banks says it can offer me a better rate if I take out at least $25k for at least three months. So even though I only think I need maybe $10k, I do it to get a better rate. Maybe the bank is hoping I find something else to spend that money on, or increase the size of my improvement. I understand that a home equity loan is backed by the house (so its not a perfect example), but that’s a case in which I’m just borrowing some money w/o an exact set of things to spend it on in advance.

    Another way to look at things in in the big picture sense: Imagine we go back to the middle of the housing bubble expansion: Banks loan money for people to buy houses. Banks continue to lower their lending standards and interest rates remain low and people see their house price go up accordingly. They think they can get rich flipping houses. As the banks lower their standards, the prices continue to rise (surprise surprise!) and more and more people are encouraged to flip houses or use their houses like an ATM. They start spending all that money on new wide screen TVs, boats, cars, Pilates lessons, etc. Where is the “backing” for the continued rise in asset prices here? The only backing was the Fed lowering interest rates, calling off the regulator watch-dogs, and the banks continuing to lower their standards.

    So overall, I agree that the money is “backed by the goods and services you buy with it” it’s also possible for the system as a whole to create a bubble, with much more money being pumped into the economy (all with offsetting debt) than is warranted, all based on a set of overvalued assets, be they tulip bulbs, houses, or internet stocks. The banks, Fed, and public work together in a positive feedback loop to create the bubble, and drive up the asset prices. When it crashes, its especially bad because all that asset wealth suddenly evaporates, but the debt is still there… it hasn’t changed a dime. Now there’s literally not enough money in the world to pay it off!

  • Romeo Fayette

    I’ve read that Keen piece before, but it still doesn’t answer my two, crucial questions: “answer the question of where the 10% comes from and/or by what mechanism the Fed increase the monetary base… then I think we’ll be done with this exercise.” (

    I do wonder what Cullen has to say in this regard, because it seems like we’re missing a crucial cog to the machine, and everyone from Keen to Fullwiler say the cog is there, but they don’t disclose its serial number–so to speak…?

  • Tom Brown

    Romeo, I wrote to Scott again with a link to this piece and your specific question. Hopefully he’ll take a look. Also, try asking Steve Keen. He’s pretty good about getting back to you in a timely fashion. You can post comments on, and he’ll often respond.

    Until then, I guess I’ll stick with my cluster of theories regarding times when demand for credit and general optimism is high, and when asset prices are rising. In these times:

    1) The banks under go the “pain” (penalties/frowns) of taking unending discount window loans from the Fed because it’s worth it…. and it only represents a fraction of the credit they extend (which is making them money).

    2) In these times standards are loosened (with regulators, the Fed, banks, investors), and borrowers are eager (and reckless). In these times “AAA” CDOs can be created from rising assets, [this is where my story line gets murky] and then traded for bank stocks bonds, etc… whatever it takes to get “investor grade” collateral put up for less painful Fed loans.

    3) Fed “seasonal borrowing” and “special programs” (whatever those are) become more numerous to meet the demands for credit.

    I think the answer to this question may be complex. I imagine that has to do with loan debt being bought and resold and insured and rated… a process which makes inherently risky assets appear to be safer than they are, and thus “collateral worthy” in the Fed’s eyes. This is helped by a sympathetic Fed which encourages people to regard their houses (for example) as “ATMs” which Greenspan famously did.

    Also, if you look at Scott’s exact words above, he really only states that the Fed doesn’t like to “lend to cover payment settlement like this” even if it’s collateralized. Perhaps collateralized lending for the purpose of meeting reserve requirements is NOT penalized. In our example the net $1000 borrowed from the Fed discount window was specifically to meet reserve requirements. No funds were borrowed from the Fed to cover the Fed overdraft (of course it would surprise me if the overdraft weren’t penalized… but then again, in the real world, the bank could probably avoid the overdraft since there are many banks to borrow reserves from… so ultimately the $1000 is really all that’s needed from the Fed).

  • Tom Brown

    Just an experiment:

    row 1, cell 1
    row 1, cell 2

    row 2, cell 1
    row 2, cell 2

  • Romeo Fayette

    I see what you’re doing here, haha. Get ready for some table porn!

  • Tom Brown

    Ha… you’re on to me! Obviously it didn’t work though… each row got divided in two. I tried the example here:

  • Tom Brown


    I got a quick turn around from Scott with this last email wherein I rephrased your question (and gave him a link to your original). I admit I don’t understand what he wrote yet, because (though I’ve looked it up numerous times in the past) I don’t recall what a “repo” is. Anyway, here’s Scott’s response:

    “Just quickly as I’m off to class in a minute, but note that the Fed uses open market operations–mostly repos–to manage the fed funds rate as its primary tool. That is, it will provide an overdraft during the day, but it expects that its open market operations will provide sufficient reserve balances for desired overnight holdings to keep the fed funds rate target. The seasonal lending, etc., are not the tools it would use to hit its target. At any rate, included in banks’ “desired overnight holdings” are reserve requirements and any additional excess banks desire to hold as a buffer to guard against overnight overdrafts or so that they don’t have to go to the Fed for an overnight collaterlized loan. And the Fed accommodates (again, mostly through its open market operations) because that’s what it means to set an interest rate target.”

  • Romeo Fayette

    I’ll respond with more later, as I’m off to a meeting myself, but for now…

    Repos are a centerpiece of the Fed’s tools, I didn’t know they were an everpresent mechanism though. Temporary Open Market Operations (TOMO) are the most common, overnight variety–as opposed to Permanent (POMO). In a TOMO, the Fed provides liquidity overnight via repos, for which a bank exchanges an high grade asset (Tbill/MBS/ABS) for cash from the Fed. A haircut is applied (i.e. if the Tbill is worth 100 cents, the Fed will lend 99.75 cents–effectively an interest charge in case the borrower defaults and the Fed is left only with its collateral). QE is a longer term version of this, hence POMO.

    Repos are used throughout the banking industry & beyond for liquidity. For example, mortgage REITs (mREITs) use repos to lever themselves up to the heavens; money market funds; custodians (like STT/BK); hedge funds; pension funds.

    I have to wrap my head around this though. My first reaction is to agree with Scott’s answer: Primary Dealers can just buy a bunch of Tbills at auction or package a bunch of loans into MBS/ABS, then they can pledge these assets as collateral to receive cash from a Fed repo (TOMO). Since we’ve established that a bank can acquire 90% of a new loan’s proceeds on the interbank market, I can now envision the remaining 10% coming from pledging the loan itself to the Fed in an overnight repo (TOMO).

    Read this way, I can see how the system leverages up and creates money… but I’ll have to run through a few thought exercises first. I do see a potential bottleneck as loan quality inevitable erodes and high-grade assets are harder to find/make/securitize in meeting the Fed’s high grade standard for repo collateral. A lot of MBS/ABS is just a bunch of crap loans combine in a pool that gets a AAA rating only because its tranched or overcollateralized–if face value is $100, then $110 worth of BB+ is as good as $100 worth of AAA.

    More to come…

  • Tom Brown

    Wow, thanks! A lot of info there. Looking forward to your further postings.

  • Tom Brown

    BTW, do you work in the “financial services industry?” You seemed to pull out some pretty detailed knowledge there.

  • Romeo Fayette

    Yes, I’m in financial services. I run a group that serves a few families, foundations & endowments. I’m a portfolio manager (investment), but our group does a significant amount of lending & banking for our clients, so I’m pretty involved with it all since I’m a partner.

    Funny, my horoscope today:
    Your curiosity might be getting the better of you, right now–your quest for knowing all the answers all the time has led you down a path that just might be a dead-end street. So today, why ask why? Probing into every little mystery will only give the mystery power over you. So don’t try to figure out why things are happening the way they are happening, don’t try to get to the bottom of the new juicy gossip. Just roll with the punches and try to get comfortable with life’s ambiguities.

  • Tom Brown

    Funny… well thanks for the great back and forth. It was illuminating!

  • Romeo Fayette

    Yes, I enjoyed too. I hope Cullen does a post on this subject matter in the near term, because it’s an important part of the foundation of modern economic thought, MMT, MR. We really got some dirt under our fingernails here!

  • JasonH

    becauses businesses cut production & layoff more people to reduce costs due to less revunue (less consumer spending or less spending from gov spending) –the cut in production reduces supply of goods, which keeps prices higher/higher

    ie, the US used to produce & sell 16 million cars per year before 2008 –50% of all new car sales are financed by autoloans

    when the 2008 finacial crisis hit, the banks severely reduced lending (aka money creation).. which resulted in only 8 milion new cars being sold… the 50% drop in income in the auto industry resulted in mililons being laid off (auto companies, dealerships, & the resturants/other businesses that depended on those workers spending) & car makers reduced production, laid off workers, closed down factories or furloughed them

    thus, car prices remained about the same due the reduced production/lower supply of cars even though only 8 million cars were sold (and milions more laid off).

    during the weak recovery, car sales slowly went back up to 12-14 milion per year as banks resumed creating money to fund autoloans & gov spending helped increase spending power to those with gov contracts/jobs

    prices remained about the same for cars despite the 50% increase in sales/demand(spending) because carmakers just increased production/supply of cars by hiring thousands more workers

    thus, increased stimulus/spending stimulates increased production & hiring as long as there is no monopoly or supply shock/shortage

    inflation is more the result of shortages or monopolies (OPEC)

    the best way to fight inflation as well as employ people is to create money (whether from private or gov doesnt’ really matter) to fund increased production of energy, food, material goods, etc

    –this was done under FDR & Eisenhower to give electricity to 90% of America outside the cities, booming agricultural/rural/suburban production , doubling medical schools, hospitals, etc

  • Andrea Malagoli

    Steven Keen’s ideas are inspired by Minsky’s work on the Financial Instability Hypothesis. It was Minsky who actually realized that not only banks can create money through esoteric structures, but that the shadow banking system (although he did not know it as such) can do so with fewer constraints than banks, and that this money creation mechanism is at the core of the intrinsic instability of a financially driven economy.

    This very basic idea is what makes Neoclassical Economics worthless, and what shows Minsky’s prescient work. Sure enough, Minsky never got the Noble Prize in Economics, and could never publish in the major economics journals …

  • Cullen Roche

    The concept of endogenous money existed well before Minsky…’s nothing new. some economists and theories seem to think they created this concept…..