Wednesday, August 14, 2013

Banking Example #11: Possible Macro Balance Sheets

In this post I am trying to capture a large set of possible balance sheets here in a simplified world consisting of just four basic entities. I'm making many simplifications (of course), but I do think it gives some insight into exactly what are some possibilities via placing simple formulas in each balance sheet cell rather than concrete numbers like I usually do.

This world consists of just four entities, two of which represent aggregates: The Treasury (Tsy), the central bank (CB), the commercial banking sector (banks), and the non-bank private sector (public). The public represents all non-bank businesses, private organizations, and individuals making up the private sector economy. This world does NOT include foreign exchange, government sponsored enterprises (GSEs), or international organizations holding CB deposits (e.g. the international monetary fund (IMF), or foreign central banks). It also does not include shadow banking or concepts such as "rehypothecation." The basic assumptions are as follows:
  1. Banks don't hold inventories of vault cash
  2. No reserve or capital requirements on the banks
  3. No coins or US notes (reserve notes are assumed as the only cash)
Now assume that the Tsy has sold $T of Tsy debt (t-debt) to the public, banks and Fed (Tsy debt not sold to one of these three entities should be excluded here: see Example #11.2 for further discussion), the CB has purchased $F of this debt, the banks have purchased another $B of it, and the public has purchased the remaining $(T-F-B) of it. Also assume that the public has taken out $L in bank loans and that they've withdrawn $C in bank deposits in the form of cash (physical bank notes). Ut is the unspent balance in the Tsy Fed deposit (the TGA). D represents the difference between the public's outstanding loan balance and the public's bank deposits + cash: i.e. it is the net total of payments to the banks by the public for interest, points, fees and service charges net of interest, employee salaries, shareholder dividends, and other payments by the banks to the public (electric bills, rent, office supplies, etc). I break my example balance sheets up into two cases. In the first, the amount of cash (C) is less than or equal to the amount of t-debt held by the CB (F) minus the unspent Tsy balance (Ut). In this case excess reserves (ER) are > 0 (the quantitative easing (QE) case). In the other case, C+Ut is greater than F but less than or equal to L+B+F-D (the amount of bank deposits held by the public if there was no cash in circulation). This is the ER = 0 or non-QE case. You can also assume that all balance sheets started off initially clear (no assets or liabilities), and that any ordering of events was used to arrive at the following balance sheet states, including Tsy auctions, lending and buying or selling of after market t-debt by the various entities. And of course, T, F, B, L, C and Ut are all non-negative. D can be either positive or negative, but must obey D < L+B+F-C-Ut. Should further net fees to banks be paid once D = L+B+F-C-Ut, it can be assumed those are incorporated into L. Note that Ut > T is allowed* (i.e. Tsy is running a surplus) however, this is rare, and it's easier in general to assume that Tsy spends every dollar it gets and thus that Ut = 0. You might wonder here, "What about mortgages?" Well certainly the Fed can buy mortgage backed securities (MBSs) so that's important to include. Also there's a government sponsored enterprise (GSE) component and a foreign component that I'm ignoring here. All that will be addressed in a subsequent post (and part of it is already!).

Independent Variables
Name  Range  Description
T 0 < T  Total Tsy debt outstanding**
B 0 < B < T-F  Tsy debt held by banks
F 0 < F < T-B  Tsy debt held by the central bank
L 0 < L  Bank loans outstanding to public
C  0 < C < L+B+F-Ut-D   Cash in circulation in public
Ut 0 < Ut < L+B+F-C-D  Unspent Tsy funds (TGA balance)
D D < L+B+F-C-Ut  Undistributed bank equity

Case 1: Excess Reserves: (ER > $0 or C+Ut < F)

Assets Liabilities
$Ut Fed deposit (TGA) $T t-debt
Negative Equity Equity
$(T-Ut) -----------

Assets Liabilities
$F t-debt $(F-C-Ut) reserves (Fed deposit for banks)
-------------- $Ut TGA (Fed deposit for Tsy)
-------------- $C cash
Total Assets Total Liabilities
$F $F

Assets Liabilities
$L loans $(L+B+F-C-Ut-D) deposits for public
$B t-debt ---------------------------------------
$(F-C-Ut) reserves ---------------------------------------
Total Assets Total Liabilities
$(L+B+F-C-Ut) $(L+B+F-C-Ut-D)
Negative Equity Equity
-------------------- $D

Assets Liabilities
$(L+B+F-C-Ut-D) deposits $L borrowing from banks
$(T-B-F) t-debt --------------------------
$C cash --------------------------
Total Assets Total Liabilities
$(T+L-Ut-D) $L
Negative Equity Equity
----------------------------- $(T-Ut-D)

Case 2:  No Excess Reserves: (ER = $0 or F < C+Ut < L+B+F-D)

Note: Only the CB and Banks balance sheets change for this case:

Assets Liabilities
$F t-debt $Ut TGA (Fed deposit for Tsy)
$(Ut+C-F) reserve loans to banks $C cash
Total Assets Total Liabilities
$(Ut+C) $(Ut+C)

Assets Liabilities
$L loans to public $(L+B+F-C-Ut-D) deposits for public
$B t-debt $(Ut+C-F) reserve borrowings
Total Assets Total Liabilities
$(L+B) $(L+B-D)
Negative Equity Equity
------------------ $D

Notice that the only change between Case 1 and Case 2 is on the CB and Banks' balance sheets: the Tsy and public balance sheets were unchanged. Also notice that in both cases the CB has no equity, while the Tsy has negative equity of $(T-Ut) and the public and banks have a combined positive equity of $(T-Ut).

For simplicity, in the rest of  the discussion below assume that Tsy spends every dollar it gets (not an outlandish assumption!) immediately and thus the unspent Tsy balance is zero (i.e. Ut = 0). This is not required but makes it easier. Also assume that D = 0 (i.e. all the banks' equity has been paid out, e.g. distributed to shareholder, etc.), again for simplicity.

For a given level of independent variables T and L, the public's money (bank deposits and cash) can vary linearly between L and L+B+F while the public's t-debt simultaneously varies linearly between T and T-B-F, depending on the amount of t-debt allocated between the CB and banks on one hand, and the public on the other.

How can this be used? Here's an example: A simple demonstration that the statement "banks lend out excess reserves" (ER) is at bit misleading. While true that cash advances do occur (C and L increasing simultaneously) I demonstrate here that what's important is that bank lending can lead to an increase in the stock of the public's money, but it has nothing to do with reserves, so the only important part of that sentence is "banks lend," the "reserves" part doesn't have any macro significance. Since my assumption is reserve requirements = 0%, all bank reserves are excess: meaning we're under "Case 1" with ER > 0. From the sheets under Case 1:

banks reserves = F - C = ER

So what happens to the non-bank private sector's (the "public's") stock of money (bank deposits + cash) if:
  1. F goes down: This results in the public's stock of money going down, $ for $.
  2. C goes up: This results in the public's stock of money staying the same.
All else (i.e. other independent variables) being equal (unchanged), in both cases. Thus whatever we twiddle here to make ER go down, the result is that the public's stock of money either goes down or stays the same. It's certainly true that both C and L can go up simultaneously, thus both increasing the stock of money and decreasing the stock of excess reserves, however think about how that can be accomplished: By literally ONLY loaning out cash (or more accurately immediately exchanging for cash the bank deposits which result from lending). Cash advances are a minor component of overall lending.

Reader Geoff has coded up a simplified version (w/o Ut or D) on a spreadsheet which I imported into MS Excel Web App for an interactive version below. It's still a ways from this very cool interactive tool but it's not too bad! Try it out by modifying the contents of the green cells at the top. You can also download a version using the Excel Web App black tool bar along the bottom of the spreadsheet (green Excel icon). Note: I don't do a range check on the inputs to make sure they're valid so if you see something odd, like a negative number anywhere except for Tsy's equity, or an odd looking plot, you probably are violating a range limitation on an input (see the above table of independent variables for valid ranges). If not, let me know! I may have a bug.

Finally note that if we set Ut = D = 0, and just focus on the five main independent variables, the public balance sheet looks like this:

Public (simplified)
Assets Liabilities
$(L+B+F-C) deposits $L borrowing from banks
$(T-B-F) t-debt -------------------------------
$C cash -------------------------------
Total Assets Total Liabilities
$(T+L) $L
Negative Equity Equity
----------------------- $T

With the public's equity in this case exactly equal to the Tsy's debt. It's also clear what the stock of public money is most reliant upon: private sector borrowing from banks (L), and institutional Tsy debt purchases (B and F). Cash (C) washes out as it simply represents bank deposits exchanged for paper notes: notes which the Fed provides in any case. If they didn't, our ATMs would run dry (which you may have noticed, they don't). All of this misses a HUGE part of what constitutes public investment and savings, for example real assets (such as real-estate). This post touches on some of what's missing from the above analysis in this regard. Simplifying further, but ignoring the distinction between deposits and cash (i.e. setting C = 0), we have:

Public (more simplified)
Assets Liabilities
$(L+B+F) deposits $L borrowing from banks
$(T-B-F) t-debt -------------------------------
Total Assets Total Liabilities
$(T+L) $L
Negative Equity Equity
----------------------- $T


* Ut > T means that the resulting "Negative Equity" for Tsy in both cases (which is normally represented with a positive number on the left here) would actually take on a negative value. Normally when that happens I null out "Negative Equity" and put positive equity on the right under "Equity" but it doesn't really matter too much: the balance sheets will balance with either method (and I kind of had to choose one since I can't have a balance with entries under both!). A similar note applies to D, but in this case I've nominally entered a positive expression on the "Equity" side of the banks' balance sheet. 

** T is total Tsy debt in this world, which includes that owned by the Fed, the banks, and the public. Specifically excluded here are foreign and intra-governmental holdings (such as Social Security, etc.)


  1. "Person y did NOT loan Bank A any money, and neither did Person x"

    Person y has a $95k deposit at bank A. That deposit IS a loan from person y to bank A.

    That's what 'liability' means: a debt, i.e. money owed.

    Bank A initially creates the deposit for person x when x takes out the mortgage from the bank. That deposit is a loan from person x to bank A.

    Then person x pays person y for the house. Now person y has the deposit. As such person y is now lending the money to bank A.

    If either person x or person y decides to withdraw the money from the bank, bank A has to pay up. This is why bank deposits are called short-term debts, payable on demand.

    1. phil?? Is that you Mr. Market? But wait, you're not this "phil" are you?:

      You're continuing this thread, I take it?:

      If you want to say that a loan is an exchange of IOUs... I can see that. For example, Bank A and Person x in this example:

      Can be thought to have exchaned IOUs after x gets the loan from A. In the end, Person y holds an IOU from Bank B, and nobody holds Person y's IOU because she didn't write one out, but Person x's IOU is continuing to be held by Bank A. I don't see that as y loaning anything out though... and I don't think I'm in the minority there. You'd say she's loaning to B because she hold's B's IOU. OK, fine, but I'm not going to start using that language...

      It sounds like you're just calling almost any financial asset a loan. Well, that's fine, but that's not the usual nomenclature and it will confuse people. I don't want to do that. Retired MBA educated banker, current bank blogger, and PragCap contibutor Frances Coppola certainly was not comfortable with this nomenclature you used:

      I think I'll stick with the mainstream on this one.

      But thanks for coming by and commenting!

      (I was very confused at first because of the "phil" ... I don't think you're my usual phil commenter, right?... plus you were referencing "Person y"... which I knew I didn't mention in this post)... Ha!

    2. yes I'm the usual "phil". I tried responding on pragcap but Cullen just deletes my comments, so I posted it here instead. Yes I'm continuing that thread :)

      Frances Coppola actually agrees with me in the comment you linked to. She says: "When you lend funds to the bank (make a customer deposit); the bank has a new liability for the amount of your deposit plus cash (asset) of the amount that you deposited". She describes a deposit as "lending funds to the bank".

      Here are some other quotes:

      "In reality, of course, a bank deposit is a loan to the bank (it is a liability on the bank balance sheet)."

      "in the traditional demand deposit the borrower can withdraw at a
 moment’s notice the money that he’s lent the bank (a deposit is a loan)."

      "A deposit is a loan contract between the depositor and the bank." (p.9)

    3. OK, point taken, however, let's take a look at the whole context of what you and Frances had written:

      phil (quoting Paul Sheard to start out):

      “Banks do need to hold reserves (as a liquidity buffer) against their deposits, and banks create deposits when they lend.”

      Nonsense. The deposits ARE the reserves. And those deposits are borrowed from the public. It’s a loan from me to the bank.


      Frances Coppola:

      No, Wrong side of the balance sheet. Deposits are liabilities, reserves are assets. When you lend funds to the bank (make a customer deposit); the bank has a new liability for the amount of your deposit plus cash (asset) of the amount that you deposited. Just like you, it puts that cash in the bank. Its bank is the central bank. Reserves are the central bank’s liability towards commercial banks. So it is incorrect to say that customer deposits at commercial banks are in any sense reserve assets. The cash that you deposited has morphed into reserves, but the deposit (as you correctly point out, the loan to the bank) is still a liability.


      I share France's objections to your original post (that you made in response to Paul Sheard's statements).

      But I know you're not a dope and not uninformed about how all this works, and you are persistent, so I'm sure you'll be back telling me where Frances and I go wrong. All I can say right now (in advance of your argument, whatever it's going to be) is that your comment at pragcap was incredibly misleading -- and probably super confusing for half the people there that are just struggling to know the difference between a reserve and a hole in the wall!... and I have no idea why you make it. I see nothing wrong w/ Sheard's statement there. But I'm sure you're going to tell me. Well, thanks for letting me know your true identity. I'd seen other Mr. Market comments which also blew me away... I thought Mr. Market was just SUPER confused and was relatively new to this stuff.... and thus was making rookie errors. I'll bet that was Frances' take too. Now that you've been unmasked I'll read Mr. Market in a different light ;^)

    4. I'm not Mr Market! I'm "phil". Mr Market is some other guy.

      I tried to comment as 'phil' over at pragcap but Cullen deleted my comments, so I posted here instead. Cullen has always deleted comments of mine in the past whenever he didn't like them, but now he won't let me comment on either pragcap or monetary realism at all.

      I disagree with Mr Market that "the deposits ARE the reserves". Of course that's wrong, and Frances comments are correct. However Mr Market is right that a bank deposit is a loan from the depositor to the bank. And Frances agrees with this, as can be seen from her comment. So your statement that "Person y did NOT loan Bank A any money" is incorrect.

    5. Ahhhh! that explains it! (regarding Mr. Market vs phil). That makes total sense now, because Mr. Market really did seem confused.

      Regarding the point you made: Yes, I grant you that! I did that already (above). OK, NOW I think I understand. Haha... I'll go back to thinking Mr. Market is a rookie. ;^)

    6. btw Paul Sheard's paper is basically all MMT. In the endnotes he says:

      "(4) Although the "money multiplier" view of central banking and credit creation is the dominant one, largely I would posit because its pedagogical attractiveness makes it a "dominant meme," other schools of thought have long existed in economics and have come to the fore more recently in the guise of "modern monetary theory (MMT)." See, for instance, Wynne Godley and Marc Lavoie, 2007: Monetary Economics: An Integrated Approach to Credit, Money, Income, Production and Wealth (Palgrave Macmillan); L. Randall Wray, 1998: Understanding Modern Money: The Key to Full Employment and Price Stability (Edgar Elgar); L. Randall Wray, 2012: Modern Monetary Theory: A Primer on Macroeconomics for Sovereign Monetary Systems (Palgrave Macmillan)."


    7. "all MMT"... yeah, I know. That came up yesterday on both pragcap and

    8. BTW, what do you think of this post and my "accounting identities" as you called them once (which I took to be a compliment actually!).

      Next I plan to add in:

      U = unspent Tsy balance

      and maybe

      D = difference between deposit and loan balances, representing interest, etc due to the banks.

      I'll let U > T, but of course U must be > 0

      D can be either +/- (the banks could get in a hole wrt the public).

      I'm regretting the use of the term "public" since it makes the term "public sector" confusing. I think I'll go back to "non-banks."

    9. Plus I think I need to modify my argument about why banks don't lend out reserves: I should emphasize that "lend out reserves" is misleading, because the stock of money is dependent on the "lending" part of that... and has nothing to do with the "reserves." So the sentence should just be shortened to "Banks lend." The "out reserves" doesn't affect the public's stock of money, which is the important part.

    10. Plus I'll add required reserves. Maybe I'll put those additions in a separate post to keep this one simple.

    11. ... and G for intra-governmental debt. Then, instead of U I'll use Ut (for unspent funds in the Tsy) and Ug (for unspent funds at the GSEs, SS, etc.).

    12. "what do you think of this post"

      Looks about right I think, given your initial assumptions. I'm not sure it clarifies things though! Maybe a verbal description would be clearer.

      Cash also includes coins however. When the Tsy sells coins to the Fed it gets a credit to its account; when it spends this it adds to reserve balances and bank deposits. So this is an additional amount that could be added to your example.

      "since my assumption is reserve requirements = 0%, all bank reserves are excess"

      If there are no official reserve requirements that doesn't mean that all reserves are 'excess'. Banks still need some reserves.

      "1. F goes down: This results in the public's stock of money going down, $ for $."

      Not if the Fed sells bonds to banks. Then the quantity of reserves goes down without affecting the quantity of bank deposits.

    13. Paul Sheard has something similar to your accounting identities in his paper:

      "D = GB + L - BK - GD.

      In words, total deposits in the banking system equal total government bonds (accumulated budget deficits) and total bank lending minus banknotes in circulation and government deposits at the central bank (this last term is subtracted because, to the extent that the government has a positive deposit at the central bank, it must have reduced the public's deposits by that net amount).

      If the public (or its nonbank intermediaries) holds government bonds directly (the usual case), things are a bit more complicated. Then, the above identity becomes:

      D = GBCB + GBBK + L - BK - GD."


      "D = difference between deposit and loan balances, representing interest, etc due to the banks"

      Could you explain what you mean?

    14. "In the other case, C is greater than F but less than or equal to L+B+F"

      How can C be greater than F?

    15. Paul Sheard: I didn't realize. But remember I did something just like this in the comments months ago (in a discussion with you), but w/o the cash. I put the cash in there for Sumner.

      re: D. Take say x takes a loan from A and then A charges him interest which he pays by allowing A to debit his deposit. My Example #10 covered that king of thing.

      re: C > F. Well, it could happen. Say ER = 0 and C = 0. Now the public want to exchange more of their deposits for cash. The banks can do that by erasing the public's deposits and then borrowing cash from the Fed to give to them. So the bank just replaces one liability w/ another.

    16. re: bond's to banks: Agreed, that's who I said "all else being equal" but perhaps I should make that more clear.

      re: verbal descriptions... clearer for some maybe, but I (and people that think like me) might like this better ;)
      Plus I've got tons of verbal descriptions elsewhere... and other sites.

      re: coins, yes I should put some words to that effect here, but I definitely don't want to get into the coin issue in the formula here! Maybe I'll save that for an upgraded edition. In fact I'm already regretting (a little) just adding the Ut variable. I did it because I could remove an assumption (Tsy spends every $). I might remove that from here and save it for a future post as well.

      Thanks for the feedback. Appreciate it!

    17. "clearer for some maybe" ... but I get your point. I like the formula in the BSs approach -- that's why I made this post in the 1st place!... but I can add extra words to clarify things. Any suggestions?

    18. Regarding excess reserves again. The UK has no official reserve requirement, but banks there still held reserves, prior to things like QE. According to wikipedia the reserve ratio was about 3.1% of of deposits, on average.

    19. "C goes up: This results in the public's stock of money staying the same."

      This isn't necessarily the case. For example if the government pays someone by check, they could just cash the check at a bank. Or, a bank might make a loan and the borrower might immediately withdraw the amount in cash. Or banks might buy things with cash. In each case the public (i.e. non-bank) stock of money would increase as cash in circulation outside of banks increased.

    20. I'd heard something like that from Nick Rowe regarding Canada too. And a few days ago someone on Beckworth's site gave me some additional info. Here's the ER held by US banks from 1984 to 2008 for example:

      Here's their comment (more stuff about Scotland):

      So, again, perhaps I can account for that in a future post, but I think I'll keep this one free from that detail. Thanks for the info.

    21. The govt doesn't send out checks to people. It's all electronic in the USA....

    22. "re: bond's to banks: Agreed, that's why I said "all else being equal"

      It's not a case of 'all else being equal'. If the Fed were to sell all the bonds acquired through QE to banks, the banks' excess reserves would go down but the bank deposits created as part of QE would not go down.

      I'm not entirely sure how your example 'proves' that banks don't "lend out" their reserves.

      If banks made loans and people immediately withdrew the money as cash, this would reduce bank reserves, so this would be a case of banks literally lending out their reserves.

      The important point in all this is that most people think banks are reserve constrained. So they think that if the Fed increases the quantity of reserves, banks will be able to lend more. The reality is that banks aren't normally reserve constrained because they can borrow reserves at the fed funds rate or discount window rate. So when the Fed increases excess reserves through QE, this doesn't increase the quantity of reserves potentially available to banks, but it does reduce the fed funds rate, unless the Fed pays interest on reserves.

    23. "The govt doesn't send out checks to people".

      As of March 2013 benefits are no longer paid with checks, except for benefits received by some seniors. But the government can still use checks to make other sorts of payments.

    24. "It's not a case of 'all else being equal'. If the Fed were to sell all the bonds acquired through QE to banks, the banks' excess reserves would go down but the bank deposits created as part of QE would not go down."

      All the independent variables being equal... NOT all the expressions dependent on those variables!

      "I'm not entirely sure how your example 'proves' that banks don't "lend out" their reserves."

      I re-worded that earlier today, perhaps you missed it.

      I say that it's misleading to say "Banks lend out reserves" because all the important macro information in that sentence is in the first two words: "Banks lend." The "reserves" part of it has no significance. That's what I now say I'm trying to prove.

    25. Well, the govt COULD do lots of things. It seems to me that most of your theory and ideas are based on this COULD world and not the DOES world. But whatever. We both know where this conversation heads (removes gun from mouth, bangs head on wall).

    26. ... but shoot! I'm glad you made me look again at it, because I had a "don't" in there which wasn't intended! I also cleaned up a couple of other minor things.

    27. phil:

      "For example if the government pays someone by check, they could just cash the check at a bank."

      That violates my assumption that Ut = $0

      "Or, a bank might make a loan and the borrower might immediately withdraw the amount in cash."

      That's a case of L increasing along with C. Let's stick to varying just one independent variable at a time, else it get's confusing.

      "Or banks might buy things with cash."

      In my model, there's no way for banks to buy things. Their equity is always 0. We could add that (actually that's on my to-do list above), but that would hardly matter. That would be an adjustment of that D term I referred to, and it would be a totally separate independent variable.

      "In each case the public (i.e. non-bank) stock of money would increase as cash in circulation outside of banks increased."

      Agreed, but you're either violating my starting assumptions or assuming more than one independent variable moving with C in each case.

    28. re: the "D" term: again, that would mean varying D along with C: i.e. changing two independent variables at once.

    29. Corrction: I wrote "In my model, there's no way for banks to buy things."

      That should say "There's no way for them to buy things other than any of the things I have here, like loans (L)."

      Anything else they bought wouldn't show up on these sheets and would push them into negative equity (the items they purchased being "off balance sheet" in this world).

    30. Cullen, do you see how valuable a teaching tool this is? Phil raises all kinds of gotcha's here, but it's easy to shoot them down because of the formulaic treatment I presented! ... everything in terms of clearly defined independent variables. How many words would be required in your typical discussion on pragcap to take care of each of these... plus some thinking too... but if everyone can see all the expressions and the clear definition of terms, there's not much to it!

      (Now watch: he'll come back with one that is a true "gotcha!") :(

    31. ... and shoot, I didn't even have to assume Ut = 0... it's (you guessed it!) another independent variable! So of course that would mean assuming a decrease in Ut along with an increase in C.

    32. "most of your theory and ideas are based on this COULD world and not the DOES world"

      No they are not.

      Regarding checks, the government still makes all sorts of payments with checks, but it has stopped using checks to make benefit payments, although some benefit recipients will still receive checks (some seniors, some in remote locations etc).

      So your statement, as usual, was factually incorrect, whereas mine wasn't.

      Your inability to distinguish between facts and nonsense you just make up inside your head is the reason why discussions with you usually lead nowhere.

    33. Tom, I agree with you that banks don't (normally) "lend out" their reserves, in the sense that reserves don't normally leave the banking system when banks make loans. Instead banks create deposits when they make loans and then use reserves to settle payments and meet cash withdrawals. However it seems to me that if someone takes a loan from a bank and immediately withdraws the full amount as cash then the bank is effectively "lending out" their reserves. If the point of your example is to prove that banks don't lend out their reserves then it seems odd to ignore this possibility.

      What I think the more knowledgeable people mean when they say that banks "lend out" their reserves, is that when banks make loans they are lending *something*, not nothing. That something is base money, because a bank deposit is a claim on base money. So even though the base money reserves usually don't actually leave the banking system, they are still being "lent out" in the sense that new claims on them are created via the creation of deposits, so the reserves could potentially be withdrawn at any time.

      Even if a bank doesn't have any reserves when the loan is made, the deposit is still a promise to pay base money on demand, either to a depositor or on their behalf. Given that a deposit is a loan from the depositor to the bank, what's happening essentially is that the bank is lending base money to the borrower, and the borrower is lending that base money right back to the bank. The base money might not even exist at that point in time, but both parties are still making promises to each other in terms of it.

      As an analogy, I might have a Rai stone in my garden that I lend to you, but you never remove the Rai stone or even touch it. It stays in my garden but you have a claim on it for the duration of the loan. I've "lent out" my Rai stone but it hasn't actually gone anywhere. Or I might not even have a Rai stone, but I have ready access to one.

    34. So phil, it seems to me that you are way underemphasizing the importance of the 1:1 ready exchangability of inside money and "base money." I could construct an argument along the same lines you do my creating almost any arbitrary definition of money. You will object and say that "base money" is not an arbitrary definition. I will agree with that in so far as it has a commonly accepted definition of some historical importance, but I would argue that the nature of our current monetary system makes the concept more arbitrary now.

      OK, so let me demonstrate what I mean by using another equally arbitrary definition. How about M0 (from wiki):

      M0: The total of all physical currency including coinage. M0 = Federal Reserve Notes + US Notes + Coins.

      So I could claim that your "base money" was simply a claim on M0. Let's take a particular part of M0, say coins:

      "Even if a bank doesn't have any coins when the loan is made, the deposit is still a promise to pay coins on demand, either to a depositor or on their behalf."

      Yes, yes, I suppose that's true, as coins are freely exchangeable with Fed deposits, reserve notes, US notes, and bank deposits, all those other forms of so-called money are really just claims on true money: coins.

      I certainly can exchange any form of "claim" money I might have for coins. A bank could do the same for any form of claim money it might have. Same goes for the Fed. It seems coins are the ONLY real money. ;)

      But of course I don't box myself in such an arbitrary corner because that would be silly! Now are there important distinctions about coins that it does make sense to take note of? Sure.. I'd say that they are bulky, metallic, and even that they are "outside" (from the point of view of the private enterprise system) all have some importance. Same goes for bank deposits... for example Tsy and the Fed don't normally (perhaps never in the latter case) deal with bank deposits *directly*. And I certainly never deal *directly* with Fed deposits. Those are important distinctions. But coins.. COINS are money to everyone! :D

    35. You win Phil. You took a marginal, almost insignificant type of transaction, and somehow validated your world view with it. You're living in lala land with your self funding govt, functional finance and Job Guarantee. The world of economics has rejected your theory. Even liberal MSM economists have rejected it. MMT has lost. The game is over. Move on. You're just wasting your time with this lala land nonsense.

      Sorry, but someone's gotta start being honest about how realistic MMT's chances of acceptance really are. You're just wasting your time supporting it. Get back to plain vanilla PKE and drop all the baggage. Help us make progress.

    36. Tom,

      what I meant, very simply, is that bank deposits are a debt of the bank. In other words a deposit signifies that the bank owes something to the depositor. That something is logically base money, right?

      Coins, notes, and reserve balances are basically interchangeable forms of base money. I can't get my hands on reserve balances directly of course, but if I have a deposit at a bank I can instruct my bank to pay reserve balances to others on my behalf. For example I can buy a bond from the Treasury by instructing my bank to pay the Treasury with reserve balances on my behalf. My bank debits my deposit account, and the central bank debits my bank's reserve account. Agreed?

      I wouldn't say it's a case of base money being the 'only real money'. For example bitcoin is a kind of free-floating currency of its own. But bank deposits aren't separate floating currencies. They are debts denominated in a particular currency. They are promises to pay a given amount of a particular currency. So if I have $100 deposit in my bank account, that means my bank owes me $100.

      As you say there is 1:1 exchangeability between deposits and base money - but only so long as a bank remains solvent!

      It is possible for a bank to become insolvent! This means that it can no longer pay its debts.

    37. Phil, you're creating your own little version of the money multiplier or reserve centric view of the monetary system. A bank does not necessarily owe you base money. It owes you your deposit. If you call on the bank to transfer your deposit then the bank must do so. It owes you the right to use your deposit so take that deposit and settle a payment with it where ever you want. If you want that settlement to take the form of base money or cash then you can do that.

      You've got this whole IOU thing from MMT which is basically your own form of the money multiplier. It's wrong.

      I don't know why you keep supporting this flawed theory. It's going nowhere. No one buys it. Not even the most liberal mainstream economists. Why bother?

      Join the rest of us PKers and get the core pieces of PKE through to the public. Drop the nonsense and "new paradigm" views. No one needs it.

    38. No it's not a "little version of the money multiplier".

      "It owes you your deposit".

      That's meaningless and incorrect.

      A deposit is a bank debt, it represents something owed by the bank. The bank doesn't owe you its own debt. That's illogical. Logic isn't your strong point is it.

      No it's not a form of the money multiplier.

      You clearly never understood the theory and still don't. Either that or everything you write about it which is completely wrong is just a cynical attempt to misinform.

      The made-up nonsense you write is not 'core PKE'.

      The stuff you write which makes sense, you got from someone else. Most of it you got from reading MMT.

      The stuff you write which is nonsense, you made up yourself. The rest of it is mainly just vague talk and rhetoric.

    39. The bank is liable to give you your deposit. That's why it's referred to as a bank liability. A bank is liable to do with your deposit what you tell it to do, whether that means giving you cash or just transferring it next door. Banks run a payment system. You might want to look into how that works.

      Other than that your comment is just the ranting of a bitter nasty man who obviously made a bad career bet on MMT and now that the MMT views have been rejected by most reputable people you're just lashing out in anger.

      Go be angry at someone else. I don't have time for nasty people in my life. And yes, I will start deleting your comments here from now on and on all of my sites.

      Good bye.

    40. That's rich coming from someone whose usual approach has been to shout abuse at me and accuse me of this and that every time I make a comment on your site questioning something you have written. And then banning me for disagreeing with you or just making comments you don't like.

      Of course in your own mind you are always in the right, you're well behaved, always moral, and what you write is true, fair, objective and balanced. In your mind you are always "under attack" by "nasty" people for no reason.

      The fact that you repeatedly lie about and misrepresent other people's work, write illogical gibberish and then shout abuse at people who question it, doesn't seem to impact your self-image one iota.

      I've yet to see anyone 'reputable' repeat any of the silly mantras you just make up and then declare to be the absolute truth.

    41. Stop being delusional. There is no "Phil". You are not a real person. You are some personal you created on the internet and you use various identities and names when you want to engage and lash out at someone. You have the luxury of anonymity and you hide behind it because you know you can attack and rant and say silly things without looking stupid. And if anyone attacks you or points out your error or rudeness then you have nothing to lose because there is no "Phil". There is only an anonymous person lashing out at Cullen Roche at every possible moment you can.

      You're not a real person. You're an anonymous coward using the internet to push your agenda and throw stones from behind your fake persona. I couldn't "attack" or "demean" you even if I wanted to because there is no you to attack.

      Me, on the other hand - what you see is what you get. I write with my real name. I fully admit making mistakes at times. I fully admit that my work is a work in progress. I fully admit that I don't have all the answers. I try my best to be nice to people, but that's often hard to do when you have an army of anonymous MMT cowards throwing stones at you. You people are tireless in your rage and promotion. You are infinitely worse than the Austrians and now everyone in economics circles is seeing it. Noah Smith has called you out on it, Paul Krugman won't even respond to MMT comments and even Marc Lavoie has called MMTers out for it.

      But this is exactly what you want. You hide behind the veil of anonymity hoping to attack Cullen Roche, a real life person. You aren't attacking my ideas. You are trying to tear down the actual person. That's not only cowardly, but it's the sort of approach that is causing people to shun MMT and all its followers because they see most of you as nothing more than bitter angry little fringe extremists.

      So take your rants and hatred elsewhere. I don't have time for your anonymous hate filled attacks.

  2. Phil, I delete your comments because I've become tired of your repeated promotion of MMT. You're doing the same thing here. You only comment when it enhances the agenda you're promoting. That's fine, but I didn't create my websites so MMTers could promote their agenda.

    It's nothing personal and I don't see why you can't understand my reasoning there. I don't care if you promote your agenda. Just don't do it on my sites.

    Of course, this is Tom's site so he can deal with you however he wants, but please try to respect my sites.


    1. My comment that you deleted wasn't about MMT so that's obviously not true.

      You deleted my comments at 'monetary realism' because I dared to ask JKH a simple question about MR.

      You deleted my comments in the past when I pointed out contradictions in your 'understanding the monetary system' text.

      You frequently misrepresent MMT in your comments, and then delete the comments of people who correct you.

      You're always promoting your "monetary realism" agenda, and you simply don't like it when people question the nonsense that you write.

    2. phil, could you email me? Thanks.

  3. Phil,

    I spend thousands of dollars a year to host a lot of websites. I've put a lot of time and effort into those sites and I try my best to provide an insightful and educational forum for people. Yes, I have my own ideas obviously and I think they're right. And I use my websites to "promote" those ideas. What I won't allow is for people like you to use the forum I provide and pay for, to undermine my efforts. If you want to pay for your own websites, spend years building the audience up and then promoting those ideas there then great! I am not stopping you and I certainly won't try to use your comments section to promote my ideas. But I know you, I know your agenda, I know your personality and I won't allow you to comment on my sites. If you can't respect that then perfect. It only validates my thinking.


  4. Hey phil, take a look at this:

    It's Vincent Cate's hyperinflation blog. He has two other posts up that greg and I were commenting on. First off there's his "Hyperinflation FAQ" that he's always trying to get people to read:

    And then there's a special hyperinflation debate page he put up:

    But greg and I zeroed in on some banking assertions he was making, which caused him to put up that 1st link about "Honest Banking." I think Vincent is deluded about hyperinflation and confused about how banking works, but he enjoys a good debate and likes to keep the tone civil. He's also pretty fair minded... he'll add to posts as he absorbs the arguments he reads in the comments.

    Take a look at that banking page. I think it's just brimming over with problems... and you can see some evidence of him changing his mind about a few things in the updates at the bottom. I'm pretty confident that given enough time I could get him to come around to some more informed views of how our monetary system works. I concentrated on that because that's what I was most comfortable with. It got to be too much of a time suck though, and I bowed out for a bit, but you might find it interesting to debate him. You might be able to convert him into an MMTer! I think that would be an improvement over where he's at now... so long as you don't make him take the blood oath and wear the purple robes and all that! :D

  5. Hi Tom, I've seen you commenting at Howfiatdies a lot and you seem very knowledgeble. I have a small blog about stock trading and I am very interested in the effects of QE on stocks and bonds. If you have anything to input on this subject please take a look at my thoughts on QE. I would be interested in any thoughts on the two questions, 1. Do QE cause stocks to rise in price? and 2. What effect do QE have on bonds? Thanks/Dan

  6. Storpappa, sorry if I deceived you into thinking I'm knowledgeable about investing. :D I'm really not! I enjoy reading about the more theoretical aspects of macro and of course banking operations. So please take my response with a grain of salt! I read the post in your link above. I think it sounds very good! I would suggest you look into a couple of issues. The first is that I believe the primary bond seller in QE is not the banks. To my understanding banks (or the primary dealers (PDs) associated with them) must act as an intermediary in trading with the Fed, but they are not necessarily the primary sellers on the open market. Thus the result looks like what I've sketched out here in my Example 4 and Example 4.1 (Case 2) more than Example 4.1 (Case 1):

    In other words it's not just bank reserves that increase, but also bank deposits. Banks only hold a smaller % of total Tsy debt not held by social security (and other intra-gov holders), foreign CBs, and the Fed, and thus they only account for a small % of sales to the Fed on the open market:

    1. Now regarding your two questions:

      1. Does QE cause stocks to rise in price. I think it does, but it's not at all clear to me. I'm very interested in the story the Market Monetarists (MMists) tell about QE. They are primarily interested in getting the Fed to adopt a better target and to do so more effectively. They believe a practical target to be the expected nominal GDP level (NGDP). NGDP growth = inflation growth + real GDP growth. They support QE, but think it is a somewhat unclear half measure.... but nonetheless they believe it should be inflationary AND promote GDP growth (i.e. NGDP growth). There is some evidence for this. I saw you linked to JP Koning... I think he's a great source on the theory of this. I would read his current articles on the convenience yield, the hot potato effect (HPE), and "forward guidance" ... I'll point you to the HPE one here, but the others were written close in time:

      Also JP left an interesting comment with linked charts at pragcap illustrating how the 1st few milliseconds after the Fed announcement of "no taper" perhaps showed the biggest effect:

      But I think it's good to be very skeptical of QE's efficacy as well. Cullen has a lot on articles on that... I won't bother digging them up... just use his search box at the bottom or Google... or just check his recent articles.

      2. Bonds: This is a super interesting question. It's not even clear to me what should happen to bond prices in theory! You might argue that they should rise (and thus the yields drop) since the Fed should be driving their price up by purchasing more. However, also from a theoretical view point, QE should be raising inflation expectations, and thus be causing yields to rise, thus lowering bond prices. The Seeking Alpha story you link to has a chart supporting this latter theoretical outcome (as you know!). Again I recommend Cullen... but it's also fun to see how the reaction of the bond market causes some confusion on the part of one of NGDP level targeting's (NGDPLT) biggest MMist boosters (Scott Sumner):

      So I'm not even going to hazard to guess there! ... however, if the MMists are generally correct, and the Fed were to adopt a more explicit goal with their QE purchases (e.g. an credible inflation target, especially if higher (say 3%), NGDPLT, or some unemployment goal) ... and do what's required to hit that goal, it's quite possible that we'd see something happen just like the MMists predict: higher inflation, higher stock prices, and higher bond yields, particularly long term (i.e. steeper yield curve). The short end of the yield curve is going to be pinned to the overnight rate, which will continue to be 0% (really the IOR rate, which is 0.25%) as long as excess reserves > 0. Cullen believes that the Fed will have a difficult time setting and hitting more aggressive goals (e.g. 5% NGDPLT) w/o some assistance from the fiscal side (i.e. gov spending or tax cuts). Also note that NGDPLT is an unproven policy: no central bank has explicitly tried it to my knowledge, whereas inflation targeting is fairly conventional (even though conventional Fed tools (FFR targeting) are not effective at the ZLB).

      That's all I got! I hope some of that was useful!

    2. Tom, thanks a lot for taking your time to answer. It was an interesting read. I've also found that the more knowledgeble people I find on this subject the less they are sure of what they believe. :)

      I was looking for your answer but I must have mixed up under which post I wrote my question so when I didn't find my comment I thought it had been deleted. And now I found your answer through google! Thanks again.

  7. A nice tool. I'm not happy with the tsy emits, cb buys tho. That will increase reserves in the table which I'm not sure is what should be happening?

    1. "nice tool" thanks. Can you elaborate a bit on your complaint though? I'm not sure I understand. The CB doesn't HAVE to buy when the Tsy sells bonds. Those are independent variables.

    2. Not sure I totally get the tool but here is my understanding, lets forget the CB part as it wasn't the root of the complaint:

      When the Treasury issues debt the banks are obligated (PD-ship and all) to buy that using reserves (to Treasury depo), then the banks might or might not sell it to the public. That is what's happening on the page you linked ( but in the calculator above Total Tsy-debt (T) doesn't seem to work that way. Did I miss something? I'm probably just mistaken but I hope you can elaborate.

      By the way I found your site really useful. It contains a lot of information that is otherwise not easily summoned.

    3. Hi Jussi, thanks for the elaboration. My tool skips the part where the banks buy the debt first and then may or may not sell it to the non-bank public (or the Fed for that matter).

      I just assume that any combination of events have already happened: with various parties buying and selling Tsy debt. I'm more interested here in what the resulting balance sheets must look like. So total Tsy debt sold must equal that owned by the Fed, the banks, and the non-banks. That's expressed as

      T = F + B + (T-F-B)

      That's all I'm doing.

      The reserves used by the banks to buy the Tsy debt may have been borrowed from the Fed, but then if the Tsy spends all the proceeds, those reserves can be paid off again. To the extent that Tsy funds are unspent (Ut > 0), then the reserve borrowing (and Fed deposit liabilities) will still exist.

      Does that help at all?

      The whole purpose was to try to briefly summarize some very simple balance sheet realities. So many questions seem to go around around chasing their tails on these points, when it's simple to sit down and just right out how it has to look, regardless of the path taken to get there.

      But I definitely invite you to work it out for yourself and see if I'm correct! I definitely want to know if I made a mistake. I think you can use that econviz tool to use a particular set of steps to move from any set of balance sheets they have to one that matches a set that my tool says is legitimate. That's all my tool does really: it doesn't say anything about the history, it just tries to show if a particular set of numbers is legitimate.

    4. I'm sorry, I definitely could improve my reading and com skills.

      I think I get it now: by selling the Treasury will reduce the amount of reserves but by spending (it assumes always U = 0) it converts reserve money to bank deposits which will put the reserves drained back into the CB (autonomous factor).

      Then F - C = ER (assuming, like you said, no CB equity)


  8. Test to see if my new blogger pic appears.

  9. Hello Everybody,
    My name is Mrs Sharon Sim. I live in Singapore and i am a happy woman today? and i told my self that any lender that rescue my family from our poor situation, i will refer any person that is looking for loan to him, he gave me happiness to me and my family, i was in need of a loan of $250,000.00 to start my life all over as i am a single mother with 3 kids I met this honest and GOD fearing man loan lender that help me with a loan of $250,000.00 SG. Dollar, he is a GOD fearing man, if you are in need of loan and you will pay back the loan please contact him tell him that is Mrs Sharon, that refer you to him. contact Dr Purva Pius,via email:( Thank you.