The Supply of Dead Presidents is Endogenous

Ashwin Parameswaran makes an incredibly important point here:

“Krugman completely misses the fact that the supply of “dead presidents” is endogenous.”

Dr. Krugman has once again delved into the abyss of endogenous money.   Throughout the crisis many economists have argued that the relationship between base money and the broader money supply is temporarily broken.  As if all those reserves are being held by banks or other agents because they’re near cash equivalents in a “liquidity trap”.  Dr. Krugman says that this is even more true due to the fact that the monetary base has historically been composed of cash.

But Ashwin makes the crucial point here.  Cash levels are endogenous.  Ie, their amounts in circulation are determined by the demand of the users of private bank money. As I’ve explained before, cash comes from outside the private sector when the US Treasury gets an order from the Fed for cash notes.  But the Fed only issues these orders when the demand from their members banks increases.  Why is this?  It is because cash is merely a facilitating feature of private bank money.  In other words, you can’t even obtain cash unless have an account at a bank whose vaults are stuffed with Federal Reserve Notes in the case that you might need some for the purposes of conveniently transacting in these notes.

These ideas of “high powered money” and the “monetary base” are unfortunate terms that muddy the waters in understanding how money is created and how the monetary system works.  In a world of endogenous money where almost all money is created by banks these concepts get large facets of the workings of the monetary system wrong.  Now, I don’t doubt that monetary policy will become more effective outside of the balance sheet recession.  But the end of the balance sheet recession won’t magically bring the money multiplier back to life.  It’s dead.  It’s always been dead and we should make sure this myth remains very dead.


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:

  • LVG

    I was waiting for you to write about this one. Krugman still doens’t understand how money is endogenous. He still completely misunderstands banking. Too bad.

  • Steve Wise

    Great headline.

  • Paul

    How can Krugman possibly not understand this? I’m not that smart of an individual, but I will say MMR is the most logical description of the monetary system I have seen. How can someone as smart as Paul Krugman say this…

    “It’s true that the Fed could sterilize the impact of a rise in the monetary base by raising the interest rate it pays on reserves, thereby keeping that base from turning into currency.”

    Someone please tell me he means something else by this, because if he means this literally, I might have to give up my admiration for Dr K after this.

  • Stephen

    Several economists employed by the NY Fed have written about this “temporary” breakdown of the money multiplier due to ZIRP, IOER, etc. Are these Fed economists dead wrong?

  • LVG

    Basically, yes.

  • LVG

    It’s scary isn’t it? But these guys are all living by defunct defintions. They think the monetary base is comprised of cash and reserves. So he thinks the Fed makes all the cash or something like that. And that they choose how much cash gets distributed. It’s bizarre that he doesn’t know this, but it all goes back to not understanding banking.

  • Steve W

    As an aside, there’s also the trend towards “the end of cash”, meaning carrying and using paper money. I believe there was an article about that in Barron’s or the WSJ the other day. Seems more and more of us just pay for stuff with credit cards or debit cards, then pay those (credit card) bills online. It won’t be much longer until it will be common to use a smart phone to “pay” for most anything, even junk food and soda out of vending machines.

    As to Krugman, I have not been reading his columns for that long. I don’t much care for his political rants. Every now and then he drifts closer to grasping MR, then he goes off on another tangent like the one mentioned in this article.

  • Stephen

    LVG — I assume that you are familiar with this NY Fed paper from July 2009:

    So according to MR, the thesis regarding the money multiplier in this paper should be dismissed?

  • Pierce Inverarity

    With regards to reserves, yes.

  • Johnny Evers

    Very confusing.
    Why have reserves grown so dramatically? Who ‘owns’ that money?

  • LVG

    Reserve levels are controlled by the Fed in the process of enacting monetary policy. They are liabilities of the fed and assets of the banks.

  • LVG

    Yes, that paper’s description of the money multiplier and why it’s broken, is wrong.

  • Johnny Evers

    Assets of the banks, or assets of people or institutions who hold money at the banks?
    If the Fed controls them, why can’t they get those reserves out into the system? Surely the Fed doesn’t want this?

  • Paul

    It’s absolutely crazy. Like I can’t even understand the logic behind it? Is he saying converting reserves to cash means banks somehow have more money or something? It’s such a puzzling statement on his part that I find it hard to believe much of what he says now. He has to have some other reasoning behind this comment.

  • Tom Brown

    Can anybody help me out with this? A question about how banks’ pay their bills:

    Say Bank A buys donuts for $10 for its employees. The donut shop holds a deposit account at Bank B. Both banks’ A and B have plenty of excess reserves (and capital), and it’s been a slow week and they’ve had no other transactions. After the payment clears, what do the banks’ balance sheets look like?

    Bank A’s reserves have decreased by $10, Bank B’s reserves have increased by $10, and Bank B’s deposit for the donut shop (a bank liability) has also increased by $10. Is that correct?

  • Stephen

    OK, thanks for the confirmation. I figured it was wrong, but a little validation never hurts.

    One more question — can’t the increase in bank deposits relative to bank loans over the past 4 years be attributed to government deficit spending? When treasuries are purchased at auction by PDs, the Treasury’s TT&L accounts and ultimately customer deposits (as treasury spends) increase.

    The liability side (deposits) of bank balance sheets increase as a result of every treasury auction, correct? So it’s not QE (and the creation of excess reserves) that has increased bank deposits, it’s the tremendous fiscal deficits, because QE is simply swapping reserves for treasuries which were already there.


  • Cullen Needs Help

    Cullen manages to acknowledge that which he does in this article, yet in his other article today, mounts a mutually exclusive attack on Jim Grant’s (valid) point that central banking (at least in the modern era under Greenspan & Bernanke) = central planning.

  • Cullen Roche

    I didn’t “attack” Grant’s question. I simply answered it. Why are you being so sensitive and rude in your comments?

  • Pierce Inverarity

    Assets of the banks, and shareholders of the banks. Not assets of you or I that have deposits at the banks.

  • Paul

    The reserves can’t “get out to the system”. Getting more excess reserves does not give banks “money” to lend people. Banks simply create money by creating deposits. But in order for them to do this they need real people and real businesses who want to take on more debt and who meet their lending requirements. So the Fed can’t force people to take on debt, they can simply affect the rate at which we borrow.

  • Paul

    Oh, that reply was for Johnny.

  • RN

    Sure, “Paul Krugman doesn’t understand banking”.

    Just wow. This blog has lost all credibility for me.

  • Paul

    Did you actually read Krugman’s article? There was nothing in it that made logical sense other than him admitting the money multiplier is misleading. His rant about these reserves being converted to cash makes absolutely no sense.

  • Tom Brown

    I was trying to locate a quote from Cullen that would partially answer the 2nd part of your question… but I didn’t, so I’ll give it a shot:

    Deficit spending: Not sure, but I don’t see offhand why that wouldn’t contribute to the gap.

    You wrote: “The liability side (deposits) of bank balance sheets increase as a result of every treasury auction, correct?” I think practically speaking that is true.

    You wrote: “So it’s not QE (and the creation of excess reserves) that has increased bank deposits, it’s the tremendous fiscal deficits, because QE is simply swapping reserves for treasuries which were already there.” — For this, I believe that QE can increase the liabilities (deposits) on a bank’s balance sheet, and in particular the balance sheet of all the banks considered as a whole. The reason is that the bank can serve as an intermediary to a non-bank for a QE Fed treasury purchase (as well as directly sell the treasury to the Fed itself). (This is the part that Cullen’s quote would have been useful for). Basically the bank sells the non-bank’s treasury to the Fed for it and then the non-bank presumable deposits the proceeds of the sale into a bank, so the all the banks’ composite deposit liabilities increase as a whole.

    Hopefully that was useful. That last bit about the bank serving as intermediary during QE was news to me.

    It seems like you may know something about the Treasury’s auction process. I asked Cullen a question about that, outlining what I supposed was the mechanism involving TT&L accounts. He responded with a quote form a NY Fed official outlining the typical process (involving “repos”), but this process did not mention TT&L accounts. Here’s Cullen’s answer with the official’s quote:

    Now, can you tell ME how TT&L accounts fit into the typical auction scenario outlined here by the NY Fed official (if they do)? Thanks!

    BTW, if you look at the “Choose Operation” menu of the following tool, you can set it to “Quantitative Easing (Variation 1 – Households Sell)” to see the effect of non-bank QE treasury sales to the Fed on all the important players:

  • Tom Brown

    I mostly agree with what you say, but one question remains in my mind: What about when banks purchase things for themselves? I have an example involving donuts here:

    So, if my idea about the balance sheets is correct, then doesn’t that amount to Bank A using it’s reserves to purchase the donuts? If I’m wrong, where did I go wrong?

  • hangemhi

    RN, if this is news to you, then I doubt you’ve spent much, if any, time reading this blog.

  • Tom Brown

    Just currious, is “hangemhi” a name? I knew a Persian woman in college whose name was Hengameh (HENG-GA-MAY), and every time I see your handle it reminds me of her.

  • http://None Midas II

    It seems to me that bank B’s deposit for the donut shop did go up $10, but that does not also increase B’s reserves (that would be a total increase of $20) which is not part of bank B’s liability to the donut shop. Reserves are there to facilitate the exchanges between banks. The reserve money bank A used was no longer reserve now that it is credited to the donut shop’s account.

  • http://None Midas II

    Tom Brown, it seems to me that bank B’s deposit for the donut shop did go up $10, but that does not also increase B’s reserves (that would be a total increase of $20) which is not part of bank B’s liability to the donut shop. Reserves are there to facilitate exchanges between banks. The reserve money bank A used was no longer reserve now that it is credited to the donut shop’s account.

  • Tom Brown

    Midas II, thanks for your response. I disagree, however w/ your assertion. I’ll sketch it out the way I see it (w/ some arbitrary values on the balance sheets before the donut purchase):

    Before donut purchase:

    Bank A: Assets: $100 reserves. Liabilities: $0. Capital: $100.

    Bank B: Assets: $100 reserves. Liabilities $0 deposit for donut shop. Capital: $100.

    Donut Shop: Assets $0 deposit w/ Bank B. Liabilities: $0. Capital: $0.

    After donut purchase clears:

    Bank A: Assets: $90 reserves. Liabilities: $0. Capital: $90.

    Bank B: Assets: $110 reserves. Liabilities $10 deposit for donut shop. Capital: $100.

    Donut Shop: Assets $10 deposit w/ Bank B. Liabilities: $0. Capital: $10.

    I thought that would be easier than spelling it out with words.

  • Anonymous

    Not true

  • Anonymous

    That wasn’t for you, Tom, but for Paul.

  • Tom Brown

    Got it! The indentations are correct, so I wasn’t confused.

  • SS

    Someone correct me if I am wrong, but Krugman says:

    “What actually limits the growth in the money supply is the fact that a substantial part of each round of lending leaks out of the banking system, getting added to hoards of green paper bearing the faces of dead presidents.”

    In normal times there isn’t high level of excess reserves. But cash withdrawals don’t constrain lending. If the cash is not deposited back in the banking system then banks, in the aggregate, will have to borrow from the Fed to meet reserve requirements, no? This still doesn’t constrain lending. Banks will make loans and borrow from the Fed after the fact if they must.


  • Tom Brown

    That’s my understanding!

  • Cullen Roche

    That’s what I took it to mean as well. Maybe he’s saying something else, but it reads pretty straight forward if you ask me.

  • Ken Pashtoon

    I’m not sure where the confusion is here, reserves are just that,held against bank available deposits, fed increases reserves allowing more deposits to be lent out. Liquidity trap meaning, bank has money to lend with not enough customers to lend to,leads to interest rate cuts, decrease reserves ,the opposite, rate rise at the point where lending demand is in competition. So no one is spending reserves, only bank capital available (1R) changes. That reserve money never really hits the street and can be used to restrict lending and affecting rates. Cash is never “created” liquidity is affected, when more capital is lent,more money is in play but can easily be reversed by the fed decreasing reserves, as loans are paid back the 1R levels are returned to pre easing levels. So it is not actually printing money, just freeing up money in play. I dont see where Krugman says different. I’ll reread to see the points being made here, can some one clarify?

  • Tom Brown

    I see several potential problems with what you say, but first a question to you: Krugman writes 1-r and 1/r, with “r” being the reserve ratio, but what do you mean when you write “1R?”

  • Tom Brown

    First off, only r*(total demand deposits) in reserves must be held by the bank to meet its reserve requirements (with r ~= 0.1 typically). Excess reserves are those in excess of this amount.

    Secondly, the bank makes the loan first, and obtains the required reserves after, if it needs to. Here’s a detailed example:

    Third, you write “allowing more deposits to be lent out” but banks don’t lend out deposits. Customer deposits are bank liabilities (money they owe to customers) which they pay interest on (to their customers).

    Forth, you write “more capital is lent” … again, capital is not lent out either. It simply represents the assets in excess of liabilities of the bank.

    I agree that (for the most part) the Fed is not “printing” more money. The Treasury prints money, and the Fed distributes this cash as needed by member banks in exchange for electronic reserve deposits. The cash held at banks is also considered to be reserves. The Fed can create or destroy reserves as needed to support the inter-bank payment settlement system and to support monetary policy.

    The banks, however, do create money (in equal measure to debt) electronically in the form of customer deposits when they make loans. The banks do not print money. The loans, of course, represent a debt (liability) of the customer (borrower) to the bank and are an asset of the bank, whereas the deposits created by the loan are a liability to the bank (and an asset to the borrower).

    Here’s another recent post with a lot of information on the process, which includes the first link I already provided at the top.

    … and this is helpful too:

  • Luke

    Stephen, Cullen answered a very similar question from me this weekend in the comments section of this article ( Basically, Cullen doesn’t think money is created in the auction process (inside money in, inside money out).

    I think I agree with Cullen, but just because money isn’t created in the auction process, doesn’t mean money isn’t created somewhere else in the process. After PDs acquire bonds, they sell them to private individuals, foreigners, and banks. Does anyone know if banks create new deposits when they acquire bonds from PDs? If not, where do all of the liabilities come from to fund all of the government bonds on commercial bank balance sheets?

  • Tom Brown

    Luke, you write:

    “Does anyone know if banks create new deposits when they acquire bonds from PDs?”

    I don’t know, but that seems backwards. Banks, (in aggregate) create customer deposits when customers sell bonds to PDs acting as intermediaries (to the Fed) during a QE operation (for example)… as long as the bond sellers are putting their proceeds in banks rather than withdrawing it as cash.

    You then write: “If not, where do all of the liabilities come from to fund all of the government bonds on commercial bank balance sheets?”

    I know there are “repo” operations typically used by PDs when acquiring treasuries during a Treasury auction. This means that money markets effectively finance these bond purchases. However, I think it’s equivalent to think of a treasury purchased at a Treasury auction by a bank as a loan to the Treasury, created with the same double entry book-keeping as is done by a bank when extending a loan to an individual or business. The Treasury, however, cannot access the “deposit” created by this loan at the bank directly, and must instruct the bank to transfer the deposit to the Treasury’s Federal Reserve deposit account. In order to accomplish this, the bank must raise reserves to match the amount transferred. This operation is analogous to what happens when a bank customer transfers a deposit from one bank to another: the old bank must raise reserves in the amount of the deposit to transfer to the new bank. In the case of the Treasury requesting the transfer, the bank continues to hold the treasury it originally “purchased” as an asset, and this can be used as collateral to borrow reserves from other banks or from the Fed discount window (as a last resort) to accomplish the transfer.

    So ultimately those funds used to fund the Treasury are obtained from the intra-bank reserve loans. Ultimately the Fed is responsible for ensuring there’s enough reserves to satisfy demand. However, banks can try to attract deposits (backed by reserves) from many sources (like new customers looking for someplace to put their money). As long as the spread between the interest they receive for their treasury and they interest they must pay to the depositors is high enough, they can make money on the deal.

  • Tom Brown

    Midas II, thanks. I don’t think that’s quite right: reserves are bank assets and the deposit for the donut shop is a bank liability, so you don’t add them together (to get $20). So Bank B’s capital (asset – liabilities) did not increase at all. Assuming that the donuts were stale and thus worthless, but the bank bought them anyway, then if the donut shop started out with a clean balance sheet (no assets or liabilities), it would gain $10 of assets in the form of the deposit at Bank B from the deal. Since it has no liabilities, that’s $10 of capital for the donut shop. See my set of before and after balance sheets below you’re 2nd (nearly identical) comment.

  • Stephen

    Cullen suggested that treasury auctions are not financed via repo back in May:

    He writes:

    “The Tsy has an account at the Fed and this account must have funds in it before it can be drawn upon. The Primary Dealers are legally obligated to make a market for US government bonds at auction. They do this by crediting the Treasury Tax & Loan account with net/new ex-nihilo money. The Treasury calls on this bank account and the Fed clears the transaction. A debit to the TT&L account is matched by a credit to the Treasury’s account at the Fed. A reserve is divested from the bank account that Treasury originally makes a call on. When the Treasury spends its account at the Fed is debited and the account of a private bank is credited with a reserve. The intermediating bank credits the bank account of one of its customers (the recipient of the Treasury spending).”

    Cullen – I think an article providing clarity on these topics would be extremely beneficial. It seems to me that there is some confusion regarding both the mechanics of treasury auctions and the end result in terms of changes in both the asset/liability side of balance sheets for all parties involved (non-bank public/depository institutions/Treasury/Fed).

    Furthermore, I think it would be very helpful to compare/contrast both the mechanics and changes to balance sheets (mentioned above) resulting from treasury auctions in both a QE and non-QE world.

  • Tom Brown

    Here’s what he wrote to me yesterday:

    Thanks for your post, I’ll take a look.

  • Cullen Roche

    That’s one of the ways this process can occur. You’re conflating two different views. I’ve been crystal clear that there are several different ways bonds can be purchased. The PRIMARY way is the repo description as per the Fed quote I often post. The comment you’re referring to is a pure bank buying and not on-selling the bond. That is, this actually results in the creation of inside money via ex-nihilo money like a loan and increases the money supply. But that’s not the primary way this occurs. The primary way this all goes down is that the bank will on-sell the bond to someone else and negate the intra-day loan in the process.

  • Tom Brown

    Ah great! I was looking for the connection between what Cullen wrote yesterday about the repos and the PDs at Treasury auctions and the TT&L accounts that they depict in this tool:

    I didn’t see how I was going to reconcile them, but this looks like the missing information I was looking for!

  • Tom Brown

    Ah great Cullen! Thanks for chiming in… that resolves a lot of confusion!

  • Luke

    For what it’s worth, I’ve thought about this a little further, and even if banks expand their balance sheets by acquiring treasuries on the secondary market, depository institutions only hold $286 billion out of $11.5 trillion in net public debt. So it’s hard to argue fiscal deficits are creating a ton of money (relative to $15 trillion GDP).

  • Cullen Roche

    Think about it logically. If the mainstream were right and the govt “printed money” then we’d have been creating a 10% increase in the money supply every year for the last 5 years via budget deficits. Does anyone really believe that?

  • Mr. Market

    In one regard Krugman is – most definitely – right. But he fails to understand the entire picture.

    The money multiplier is dead and has been dead for centuries. In fact, it never existed. It debt and credit that been multiplying for – at least – decades.

  • Geoff

    Tom, it would be good to take your example a step further and analyze the aggregate impact. As far as I can tell, the amount of Bank assets (reserves) in the system remains the same at $200, but overall bank capital has declined from $200 to $190 due to the $10 expenditure on the donuts. But that $10 decline in bank capital is a $10 gain for the donut shop.

  • Tom Brown

    Hi Geoff, thanks for commenting. What you say is true, I think, provided that what I say is true. But that’s my real question here… does the Bank use it’s reserves to pay it’s bills? If my balance sheets are correct, it would seem that it does.

  • Tom Brown

    Geoff, re-reading your comment I want to state that my accounting skills (poor!) break down when dealing with the donuts before the purchase, which is why I qualified them in this paragraph above:

    as being “stale and worthless but the bank bought them anyway.” I did that because, frankly, I didn’t know how to deal with them on the donut shop’s balance sheet BEFORE the purchase (I assume they’d be an asset, but do you write down the “market value” or the value it cost to make them? I assume the market value… but then again, how does the donut shop increase its capital then? By making he donuts, not selling them?…). So you see, I didn’t really want to get into all that. I was more interested in my basic question:

    Does the bank use its reserves to pay its bills, and if not, what does it use then?

    So you’re right, bank reserves didn’t change, but bank capital did. But it’s really an artifact of my contrived donut shop balance sheet that the decline in bank capital was neatly offset exactly by a rise in the donut shop capital. If those donuts had been represented as a non-zero asset on the donut shop sheet balance sheet prior to the sale that wouldn’t have been true.

  • Geoff

    Hi Tom,

    I’m not an accountant either, nor a banking expert, but I believe reserves are like a bank account (the bank’s bank account held at the Fed). A bank can use it’s “bank account” to pay bills just like you and I can.

  • Tom Brown

    I thinking that’s correct too, but it’s a little odd. In general when a business writes another business a check, the bank moves the deposit (debiting one account and crediting the other). If the businesses have separate banks, then reserves must move as well. But when a bank writes a check… what happens? In my example above, no liability (deposit) MOVES… it’s just created! The reserves still have to move (if two banks are involved). This all makes sense to me, but it seems to be a special case of a business writing a check. BTW, if the donut shop had an account at Bank A, instead, then Bank A would simply credit their deposit by $10. It all works out…, Bank A capital still goes down by $10, and the donut shop’s goes up by $10 (and Bank B stays the same).

    In the bigger picture though, this brings up an interesting question. We state on this site over and over that banks don’t do anything with their reserves other than settle payments with them and loan them ONLY to other banks… and trade them with the Fed for treasuries or whatever during Fed operations. However, if a bank pays its bills from its reserves, what’s to prevent it from buying other stuff? …. like Greek bonds? I’m not saying it happens, I’m just wondering what prevents that (if anything)?

  • Tom Brown

    For that matter, I guess I neglected the donuts on the bank’s balance sheet AFTER the purchase! Ha!

  • Geoff


    Remember that banks are capital constrained, not reserve constrained. You could say it is their net assets that are the key. Whether a bank holds more of its assets in reserves or some other instrument like bonds makes little difference. The high amount of reserves currently in the system shows more about the asset MIX rather than asset total.

  • Geoff

    I think those crispy cremes are in stomachs rather than on any balance sheet, or perhaps down the toilet bowl in the case of Kate Moss ;)

  • Stephen

    Cullen writes:

    “Think about it logically. If the mainstream were right and the govt “printed money” then we’d have been creating a 10% increase in the money supply every year for the last 5 years via budget deficits. Does anyone really believe that?”

    It’s not that hard to believe — M2 has grown from $7.457 trillion on 12/31/07 to $10.516 trillion on 12/31/12. This is an 8.2% annualized increase in M2 money supply…

  • Stephen

    John Aziz brought up a good point about the Fed’s role in money creation:

    While it’s true that the Fed does not itself create money, it’s explicit consent is required (via increases in bank reserve supply) for private banks to create money.

    So while the Fed does not directly increase the money supply, its involvement and consent is a prerequisite for money supply growth.

    Furthermore, and as John points out, it is the Fed’s responsibility to monitor private bank credit expansion (which can only occur if demand for credit is sufficient to support that expansion) and impose credit contractions (via higher reserve requirements, refusal to supply reserves thus increasing FFR, etc.) to prevent credit bubbles (and by rules of accounting, inflated money supply as measured in bank deposits).

    So in that context, although the Fed does not create money, it permits money creation. Without the Fed’s willingness to supply reserves, private banks cannot create money over the long-term.

  • Cullen Roche

    There’s really no fed choice in the matter. If a bank chooses to make a loan the Fed must supply required reserves if they’re not available. The money supply is privatized.

  • Stephen

    Yes, however if the growth rate of loans and by extension required reserves/deposits is increasing at an unreasonable level, it is the Fed’s responsibility to reduce the amount of excess reserves in the system, exerting upwards pressure on the FFR, which in turn will reduce demand for credit — retarding (and reversing if necessary) the growth rate of the loans/required reserves/deposits supply chain.

  • Cullen Roche

    Yeah, the Fed can influence the cost of money. But I’d be careful about saying it controls the money supply. 95% of the time the Fed is there to support credit creation. Not to suffocate it.

  • bart

    So very few see the inflation that has actually occurred. Oh well…

    “There is no subtler, no surer means of overturning the existing basis of society than to debauch the currency. The process engages all the hidden forces of economic law on the side of destruction, and does it in a manner which not one man in a million can diagnose.”
    — John Maynard Keynes, “Economic consequences of the peace”- Unseen Hand, page 57

  • Stephen

    Just because they have historically always supported credit creation, does not mean that’s always the best approach.

    One wonders what US total credit market debt($55.3 trillion excluding intragovernmental federal debt as of Q3 2012, 350% of GDP) would look like if the Fed wasn’t supporting credit creation 95% of the time.

  • bart

    What about when dead presidents disappear totally and we’re 100% digital, as almost everybody wants?

  • Tom Brown

    That’s when the hackers strike and erase all the records on all the bank and central bank computers that they’ve hacked, casting the world into chaos!

  • Tom Brown

    Yes, of course… I’m being a bit ridiculous here, but *just before* consumption, they must count as an asset. Ha!

  • Tom Brown

    Geoff, yes I agree that “banks are capital constrained, not reserve constrained” when they make a loan. But I don’t have any loans going on in my example.

    “net assets” … isn’t that another word for “capital,” i.e. assets in excess of liabilities?

    You write “Whether a bank holds more of its assets in reserves or some other instrument like bonds makes little difference.” but I disagree to some extent. I think we just saw the consequences of a problem with this… where the banks held LOTS of assets in crap mortgages, but they didn’t want to have to “mark them to market” so they sat there with a high face value, but everyone new the banks were in trouble because their assets were toxic.

    So… I don’t mean to pick apart your post, I think you and I are in agreement on most of this. In fact I’m very happy that you took the time to consider my questions! Thanks! But I’m still left wanting: What are the mechanics of how a bank pays its bills? Have I got it essentially correct with my donut example?

  • Tom Brown

    Ah! OK, I think I see what you’re getting at! Sorry, a little slow: you are saying to my broader question, that its fine if the bank swaps reserves for bonds (high quality bonds) because it doesn’t affect the balance sheet. Is that it?

    The thing is, doesn’t that amount to them using their reserves to invest with?

  • Geoff

    Right. The spending power of a bank I think comes from the same place as it does for you and I, i.e. from net worth, or capital. The amount of money I have in the bank (or the amount of reserves a bank has at the Fed) doesn’t tell the whole story. For example, I normally keep around $2K in my bank account (I’d keep less but then I’d be charged exorbitant fees). Plus I have around $100 bucks in my wallet. So my contribution to the money supply is around $2,100. But my net worth, or my liquid net worth (my real spending power) is slightly higher :)

  • InvestorX

    Because the system was perverted. What you call now deposits are actually credit lines for zero interest (and zero fee as a remuneration).

    It used to be that reserves were held as a liquidity buffer to serve “deposit” withdrawals. And in a 100% reserved system there are no bank runs (unless a bank becomes underreserved somehow)

  • InvestorX

    The Fed does not print money, it prints reserves. Question is what are reserves to banks and are they money or not for the banks. In the past gold = reserves = money. Nowadays – no idea?

    Banks do not print money – they print inside money. So if you ask me they DO print money, it is only in electronic deposits, but easily convertible in dead presidents by the customers, who notice no difference.

  • John

    The news reports equate failure to increase the debt ceiling with default. I don’t understand. Cannot the government simply produce new funds (print money) by some of the operations described above and elsewhere? These funds are then used to pay the bills wothout borrowing additional amounts.

  • Tom Brown

    Let me take a crack at that one. While technically true that the Treasury (part of the gov) prints physical notes and and mints coins (cash) at the behest of the Fed (once at the Fed, the currency becomes available to distribute to banks as “reserves” in exchange for electronic reserves the banks already hold), that is a small part of the money supply. The Fed is really responsible for creating the “inside” or exogenous money. But the Fed is NOT the government. It’s a hybrid public/private independent institution. If the government wants to spend money, it needs to access funds from the Treasury’s Federal Reserve deposit account. Banks also maintain Federal Reserve deposits accounts. The Fed facilitates the movement of these reserves between deposit account holders, and can inject unlimited amounts of reserves (that it creates out of thin air) into PRIVATE bank reserve deposit accounts in exchange for other financial assets held by the banks (such as treasury bonds). So normally what happens is the Treasury is allowed (by congress) to sell bonds (treasuries) at a treasury auction. If the debt ceiling isn’t raised, then that auction can’t take place. The funds from the auction eventually find their way into the Treasury’s Fed reserve deposit account, from which they can be used by the government. The Fed can not just legally credit (increase) the Treasury’s reserve deposit account… those reserves must be transferred in/out from/to other deposit account holders (banks).

    So no, with the system we have, the government cannot just “print” the money, either physical money or electronic money, and have it available to spend.

  • Cowpoke

    Tom Thoughts?
    Suppose now that the
    central bank lends $40 directly to Firm X, and suppose that this firm holds a deposit account at
    Bank A. In making this loan, the central bank credits $40 to Bank A’s reserve account and Bank
    A, in turn, credits $40 to Firm X’s deposit account.

    it is important to keep in mind that total reserves
    in the banking system are determined almost entirely by the central bank’s actions. An
    individual bank can reduce its reserves by lending them out or using them to purchase other
    assets, but these actions do not change the total level of reserves in the banking system.

  • Cowpoke

    Tom, have you had a chance to read over this one?

  • Tom Brown

    No I haven’t… did you ask me to take a look?

  • Tom Brown

    Ah, OK now I see where you’ve asked me to. OK, I’ll think about it. Gotta go right now though… thanks!

  • Cowpoke

    Yes, can you give it a read?
    Especially Page 6&7

  • Tom Brown

    OK, haven’t read your link yet (pg 6-7, got it!), but your paragraph sounds OK, except for the part about “Central bank loans directly to Firm X” … that’s certainly not an everyday event, right? Did that actually happen in 2008/2009?

    I think I know where you stand on my basic question Cowpoke: As long as the reserves are excess, you’d say, then yes, the bank can do as it pleases with them… bet it all on black at Vegas, whatever, right?

    I’m not sure exactly what you’re getting at… I agree with your statement “An
    individual bank can reduce its reserves by lending them out or using them to purchase other
    assets, but these actions do not change the total level of reserves in the banking system.” … at least I think I agree… haven’t thought about it too much… my example above certainly is consistent with that (still $200 of reserves in the system before and after the donut purchase).

    Well, I guess what I’m trying to understand (primarily) is: is your position about the bank being able to do anything it wants with (excess) reserves really true? Or if it’s not, what are the true limitations, and who sets those?

    Thanks again!

  • Tom Brown

    Ah, OK, I finished pg 6-7. You were just taking their example…and it states plainly there on pg 6 that the Fed can lend directly to firms! Plus it says that banks can “lend out” their reserves (I think to other banks? — since the bit on pg. 7 makes it clear that making a loan to a firm does not decrease the banks reserves) or use them to purchase assets. I guess that bit still doesn’t answer my question though… are there limitations on what kind of assets? Purchasing treasuries during OMO … I knew that already… those are assets. But what about the electric bill?

    OK, I’ll read the rest when I get a chance.

  • Cowpoke

    Fed Will Lend Directly to Corporations

    “The Federal Reserve said it will bypass ailing banks and lend directly to American corporations for the first time since the Great Depression”

  • Tom Brown

    Sure, got it. I should have read pg 6-7 before commenting on your paragraph… I didn’t know you’d just taken it directly from there. Nice quote from the WSJ.

  • Tom Brown

    InvestorX, I agree w/ your assertions here. You also ask “what are reserves to banks and are they money or not for the banks.” I think the answer to that is “yes, they are money” however money with restrictions. A bank would be prohibited by regulators from spending all their reserves if that would result in their equity position (assets in excess of liabilities) diminishing into dangerous territory (e.g. negative equity). I was asking a related question recently and got a wonderful detailed answer from Joe in Accounting. You are familiar with the question (How do banks pay their bills?) since I know you commented on one version of it. Here’s Joe’s response: