Another Nail in the Money-Multiplier Coffin

It’s been awhile since my last rant about the money multiplier. I am not really feeling up to another whack at the piñata just yet, but via Not Trampis, I saw this post on the myth of the money multiplier by the estimable Barkley Rosser. Rosser discusses a recent unpublished paper by two Fed economists, Seth Carpenter and Selva Demiralp, entitled “Money, Reserves, and the Transmission of Monetary Policy: Does the Money Multiplier Exist?”

Rosser concludes his post as follows:

That Fed control over the money supply has become a phantom has been quite clear since the Minsky moment in 2008, with the Fed massively expanding its balance sheet without much resulting increase in measured money supply.  This of course has made a hash of all the people ranting about the Fed “printing money,” which presumably will lead to hyperinflation any minute (eeek!).  But the deeper story that some of us were unaware of is that apparently this disjuncture happened a long time ago.  Even so, one of our number pointed out that official Fed literature and even many Fed employees still sell the reserve base story tied to a money multiplier to the public, just as one continues to find it in the textbooks,  But apparently most of them know better, and the money multiplier became a myth a long time ago.

Here’s the abstract of the Carpenter and Demiralp paper.

With the use of nontraditional policy tools, the level of reserve balances has risen significantly in the United States since 2007. Before the financial crisis, reserve balances were roughly $20 billion whereas the level has risen well past $1 trillion. The effect of reserve balances in simple macroeconomic models often comes through the money multiplier, affecting the money supply and the amount of lending in the economy. Most models currently used for macroeconomic policy analysis, however, either exclude money or model money demand as entirely endogenous, thus precluding any causal role for money. Nevertheless, some academic research and many textbooks continue to use the money multiplier concept in discussions of money. We explore the institutional structure of the transmissions mechanism beginning with open market operations through to money and loans. We then undertake empirical analysis of the relationship among reserve balances, money, and bank lending. We use aggregate as well as bank-level data in a VAR framework and document that the mechanism does not work through the standard multiplier model or the bank lending channel. In particular, if the level of reserve balances is expected to have an impact on the economy, it seems unlikely that the standard multiplier analysis will explain the effect.

And here’s my take from 25 years ago in my book Free Banking and Monetary Reform (p. 173)

The conventional models break down the money supply into high-power money [the monetary base] and the money multiplier. The money multiplier summarizes the supposedly stable numerical relationship between high-powered money and the total stock of money. Thus an injection of reserves, by increasing high-powered money, is supposed to have a determinate multiplier effect on the stock of money. But in Tobin’s analysis, the implications of an injection of reserves were ambiguous. The result depended on how the added reserves affected interest rates and, in turn, the costs and revenues from creating deposits. It was only a legal prohibition of paying interest on deposits, which kept marginal revenue above marginal cost, that created an incentive for banks to expand whenever they acquired additional reserves.

When regulation Q was abolished, it meant lights out for the money-multiplier.


30 Responses to “Another Nail in the Money-Multiplier Coffin”

  1. 1 Greg Hill February 7, 2013 at 11:13 pm

    David, good stuff as always. You might also be interested in this:


  2. 2 Marcus Nunes February 8, 2013 at 3:40 am

    The mm would have a role if the base/reserves was “frozen”. In that case the mm would go up and down to ‘satisfy’ money demand. Maybe that would be equivalent to ‘free banking’?


  3. 3 PeterP February 8, 2013 at 3:40 am

    One day even Scott Sumner will get the memo.


  4. 4 pcle February 8, 2013 at 7:51 am

    I believe Charles Goodhart was discussing the irrelevance of the money multiplier back in the 1970s, in his book Money & Uncertainty. One more reason why he is the best monetary economist of the last 50 years by some distance.


  5. 5 Mike Sproul February 8, 2013 at 8:33 am

    Say there are 100 shares of GM stock in existence (backed by GM’s assets), and then traders issue 200 claims to GM stock, backed by their personal assets. The “GM stock multiplier” just rose from 1 to 3, but of course the GM share price is unaffected. If GM then issued another 100 shares of stock in exchange for equal-valued assets, then the GM share price would still be unaffected, and we might see 100 of those GM claims reflux to their issuers and get retired.

    The story is the same if the Fed has issued 100 paper dollars (backed by the Fed’s assets) and Wells Fargo then issues 200 checking account dollars (backed by Wells Fargo’s assets) that are each a claim to a paper dollar. The money multiplier would rise from 1 to 3 and the value of the dollar would not change. If the Fed then issued another 100 paper dollars in exchange for $100 worth of bonds, then the value of the dollar would not change, and we might see 100 checking account dollars reflux to Wells Fargo and get retired.


  6. 6 Tom Brown February 8, 2013 at 6:34 pm

    Hi David, nice piece on the money multiplier. Sorry to be a bit off topic here, but I was wondering what you thought might be the consequence to the economy and Market Monetarist theory if we were to go to an all electronic money system. Any significant effects?


  7. 7 Ilya February 9, 2013 at 1:55 pm

    But if there is no stable money multiplier then doesn’t that mean money is endogenous and the fed does not control prices and NGDP? I’m confused.


  8. 8 Will February 9, 2013 at 2:16 pm

    Hold on. I’m not the brightest, and I’m surely missing something. But isn’t there a certain tension between the Banking School view, which David seems to be endorsing here, and the Currency School view that underlies monetarism? If neither central banks’ “easing” operations (which increase the amount of bank reserves), nor their control of interest rates (which changes the relative attractiveness of holding reserves or lending them out), has a deterministic relationship to the quantity of currency circulating, than aren’t we left with the “expectations” channel alone, which would in this story depend solely on a widespread misperception (unless monetarists join me in embracing counterfeiting)?


  9. 9 Tom Brown February 9, 2013 at 9:50 pm

    Most money is endogenous (bank created). I think David and Nick Rowe would agree with that. Rowe would even agree that over a short time interval (like six weeks — the time between revisiting the target interest rate for the BoC) that this endogenous money supply is even “perfectly elastic” with respect to this fixed interest rate: Meaning the CB will provide reserves to the banking system as required to defend this fixed overnight interest rate. It’s only over a larger time frame this he disagrees with this (like 2 years or so) and that’s because what the CB actually targets is the inflation rate. I can’t tell you more because I really don’t understand this short term vs mid term time distinction that he makes (although I have tried on numerous occasions), especially when he says that over the mid term the supply of money becomes **perfectly** inelastic wrt the overnight rate, and perfectly elastic wrt the inflation rate. The best I can wrap my head around is that it may become imperfectly inelastic wrt the overnight rate because it seems to me that people’s desire to borrow funds and the inflation rate are independent (imperfectly correlated) variables.

    Regarding “lending out reserves” this doesn’t happen except to other banks. In fact the only four ways reserves can leave the private banking system (as a whole, i.e. on the composite banks’ balance sheet) are:

    1. When private non-banks withdraw them in the form of paper bills and coins. I hate the words “cash” and “currency” because people use those in ALL kinds of confusing and contradictory ways. However, “vault cash” (also paper bills and coins) at the bank is defined as reserves, and can be exchanged back to the Fed for electronic deposits (also reserves).

    2. Taxes paid by anyone (bank or non-bank) to the Federal government.

    3. Treasury bond auctions: when either bank (PDs) or non-banks (through the on-selling of the bonds or from “Treasury Direct” purchases) buy these assets from Treasury. There is a small technical matter in that in some forms of these auctions the PD will establish a “TT&L” deposit for treasury which is very similar to the bank loaning Treasury the funds (i.e. the bank simply makes an entry on the liabilities side of its balance sheet representing the Treasury’s deposit). On the asset side, instead of the usual loan papers, they hold the Treasury bond (which again is an electronic entity). The reserves don’t actually leave the banking system until the Treasury instructs the bank to transfer its TT&L deposit to Treasury’s Fed deposit account, from which the government (not counting the Fed here as part of the government) can make use of these funds. Technically the Treasury’s Fed deposit is not counted as “reserves in the banking system.”

    4. Fed OMSs (OMOs in which the Fed sells assets).

    Up till now I have not made a distinction between excess and required reserves (ERs and RRs), although I believe all four of the above practically apply just to ERs. Worth mentioning though is that banks can convert ERs into RRs through making electronic loans to private non-banks. 10% of the loan principal in ERs are converted to RRs when this happens (to meet reserve requirements). For example, on the composite bank’s balance sheet, if an individual bank loans out $100 then $10 of ERs get converted to RRs. I emphasize that I’m referring to the composite banks’ balance sheet, since any individual bank may need to borrow another bank’s ERs to affect this conversion into RRs. I’m assuming here that ALL loans made by the bank are originally “electronic” and that if a customer takes a loan, keeps the loan in his deposit account, and then withdraws the funds as paper bills and coins, this process is already covered by 1. above.

    Of course individual banks CAN diminish their ERs in many other ways: using them to pay bills, salaries, dividends, loaning them to other banks, or even to purchase assets from other private entities (bank or non-bank). As long as the recipient of these reserves keeps its deposit account at another bank, the originating bank’s ERs are diminished. It’s just that in aggregate, reserves (ER+RR) don’t leave the private banking system except for the above four ways and in addition ERs can be converted to RRs when loans are made.

    But even if there are no ERs in the private banking system, this does not prevent banks from making loans. In order for the CB to facilitate the operation of the private banking system and to clear payments (one of its main mandates) and to defend the Fed overnight target rate, it must provide the additional RRs to the banking system through overdrafts, OMOs, or though lending them through the discount window. Again Rowe would disagree w/ this last assertion when talking about longer time frames. I’d be curious as to what David thinks of this. For more information on Rowe’s views google “The supply of money is demand determined” or “Banking Mysticism.”

    Personally I can’t figure out Sumner’s views, but one of his “acolytes,” dtoh, does seem to agree. Sumner, in other posts, seems to mostly agree with dtoh’s views, however there are some differences in the way they look at things and Sumner admits that dtoh’s explanation of MM may be more “marketable.” For anyone who’s curious, here’s the start of a long interchange I had with dtoh on these subjects:

    One of the advantages of dtoh’s explanation for someone like me who’s new to MM is that is does not require a “hot potato effect” and does not ascribe any special significance to paper bills and coins.


  10. 10 Tom Brown February 9, 2013 at 9:58 pm

    I should probably qualify this last sentence and say “any special causal relationship to paper bills and coins” (i.e. he doesn’t claim paper bills and coins in circulation CAUSE an increase in AD, but they may be correlated with it). He also claims that if we were to get rid of all paper bills and coins it would not affect the workings of MM (although there would be some other effects).


  11. 11 Tom Brown February 9, 2013 at 10:25 pm

    … shoot!… and by “MM” in this last post I mean “Market Monetarism” not money multiplier. Or is that what everybody understands already when it’s MM vs mm?


  12. 12 a.concerned.reader February 10, 2013 at 5:45 am


    I would appreciate your taking the time to be clear what you claim to be a myth.

    It is not a myth that private banks can create money.

    Under present circumstances they may, or may not being doing such. In the future they may or may not be doing such.

    The Fed may or may not be able to force or prevent this behavior.

    But, their ability to do such is not a myth and, given you rush to make your point, you seem to mislead and state they they cannot create money, and that is not true.


  13. 13 Becky Hargrove February 10, 2013 at 10:51 am

    Please don’t be so quick to run to the goal line, out of frustration with that “broken transmission”! (Sleep a couple more nights on what the 2010 graphs you put together were all about, if need be). After all “broken transmissions” sometimes need to be rebuilt to better specification. Also, multipliers – public or private in nature – and whether they even exist at any moment, depend on the degree to which certain forms of economic expectation have been filled, and the present or ongoing capacity to fill them further.


  14. 14 Tom Brown February 10, 2013 at 3:15 pm


    I think you’re misreading David. I don’t think he is claiming that banks cannot create money. Did you follow the link to his other article?:

    Nick Rowe agrees that banks can create money:

    “Strangely, while I agree with the MMT guys on many points (banks do create money out of thin air) … ” – N. Rowe

    Although in that debate he goes on to say (the 2nd half of that sentence):

    “…I end up (I think) in much the same place as Paul [Krugman, in his debate with Steve Keen.]” – N. Rowe

    In Rowe’s very next article:

    He states:

    “So drawing a perfectly interest-elastic money supply curve is a reasonable approximation to reality for 6 week periods.” – N. Rowe

    So what this means is that he’s in agreement that over these 6 week periods the BoC does supply reserves to support the amount of money created by the banks “out of thin air” endogenously. Now since there are no reserve requirements in Canada, this means the BoC only needs to create, net, reserves to meet any reserve buffer desires of the private banks. I’m claiming this is an insignificant amount. Nick goes on to say, however:

    “For any longer period of time, it’s totally wrong.” – N. Rowe

    I personally do not fully understand the difference in periods of time.

    David, are you in agreement with Nick over these times periods between CB fixed overnight target rate changes?

    Again assuming that amounts of buffer ER (desired to be held by banks to avoid overdrafts, etc) are insignificant, then we have two basic situations:

    1. ER = 0. In this case the CB adjusts reserve levels through OMOs to defend its overnight target rate. The target rate must be >= IOR for this to work. In this case the arrow of causation runs from loan activity back to actions the CB must take. In the US, with the required reserve ratio approximately 10%, this means that the CB will change the reserve levels in the system up or down to match 10% of the net change in funds loaned out.

    2. ER > 0. In this case the CB can only defend its target overnight rate by setting the IOR equal to it. The CB need not change reserve levels in response to loan activity. Conversely, changes in reserve levels have little effect on loan activity. (actually I think most Market Monetarists would disagree w/ this last statement, but the mechanism is unclear to me still).

    In either case the money multiplier (mm) is dead, or at least does not function in the way usually presented.

    In case 1. above the required reserve ratio only determines what the CB must do to defend its overnight rate. In case 2. the required reserve ratio has no effect on the CB. In neither case does it affect loan activity and thus endogenous money creation.

    Now Nick Rowe says, however, that these small intervals of time (the time between CB rate changes) over which I assert the above is true are uninteresting macro economically speaking.


  15. 15 nottrampis February 10, 2013 at 5:18 pm

    As it is that paper by Palley was the next one I highlighted after this one David is writing about.

    Hopefully he will write about soon and together with the usual excellent comments here we can all get a much better understanding of the topic.


  16. 16 a.concerned.reader@ February 11, 2013 at 10:49 am

    Tim Brown

    Thank you for the support.

    Now, to the main point.

    Actual behavior of banks can be observed and measured, which is scientific. Feynman, among many

    Accordingly, whether private bank creation of money is of moment is a fact to be determined by actual observation, not arguments no different in kind than the number of angels on the head of a pin.

    However, by the mere fact that we let banks loan money and we see it as a problem when banks either loan too little or not enough, our own behavior negates the claim that bank creation of money is a myth.

    To the contrary, we can show now that bank failure to create money is a problem for the economy. We don’t have enough money in bank accounts that people intend to spend, either for consumption or investment.

    So, it seems to me that we have, at the end, a paper and a post, both appearing erudite, but neither providing knowledge of use


  17. 17 David Glasner February 11, 2013 at 2:57 pm

    Sorry these responses are so late in coming.

    Greg, Thanks both for the comment and the link which looks very interesting. I need someone to give me an overview of MMT.

    Marcus, You are identifying the key conceptual flaw in the money multiplier. The money multiplier is supposed to represent the link between the monetary base and the total money supply. With the total money supply determined you can then solve for the price level be equating demand and supply. But as your example makes clear what causes the money multiplier to change is the _demand_ for money. If your supply function depends on demand, how can you use supply and demand to solve for anything? The whole money multiplier apparatus is based on a confusion of demand concepts and supply concepts.

    PeterP, Not sure what you mean. Scott doesn’t believe in the money multiplier either.

    Pcle, He wasn’t the only one. Tobin published his classic paper “Commercial Banks as Creators of Money” in 1963. But I am not going to argue that Goodhart is not the best monetary economist of the last 50 years. He just might be, but I think that there are also others who might be.

    Mike, Nicely put.

    Tom, Can’t speak on behalf of Market Monetarist theory. As I have said elsewhere, in principle, I don’t see any fundamental difference between electronic money and paper money.

    Ilya, I think that you are expressing a view that some people (post-Keynesians, I think, among others) hold. I disagree, because I think that, as a first approximation, prices can be determined in a model simply by the stock of and the demand for base money, and the stock of base money – and here I disagree with Mike Sproul – is under the control of the monopolistic supplier of base money. The supply of bank money is endogenous, and (as a first approximation) does not affect the price level. The supply of base money is exogenous, and its supply (given the demand) determines the price level.

    Will, I am a banking school guy through and through. And you are right that the Friedmanite version of Monetarism is a derivative of the Currency School. But there are alternative ways of doing monetary theory that preserve a central role for monetary policy that are in the Banking School tradition. You can read my book where this is all spelled out in detail or my papers on classical monetary theory, especially “A Reinterpretation of Classical Monetary Theory, in the Southern Economic Journal 1985, where the basic theory was developed.

    Tom, The central bank does have to lend to the banks to supply them with the reserves they need, but the central bank gets to determine the rate at which it makes those reserves available. Just because they will supply whatever amount of reserves is demanded at the chosen interest rate does not mean that the supply of reserves is not under the control of the central bank. That seems to me a basic mistake of the endogenous money people. It’s exactly the same idea as the duality between controlling imports by a fixed quota or by a tariff. When you are not exactly sure what the demand or supply is, it is less disruptive to set an interest rate and allow demand to determine quantity or to fix a tariff and allow supply to determine quantity. But in either case the quantity is exogenously determined by the policy.

    a.concerned.reader, I don’t believe that I ever said or wrote that it is a myth that banks create money. That is certainly not what I believe. Do you have a quotation of mine that you are referring to?

    Becky, Sorry, but I am having trouble linking you comment to something that I said. Would you mind explaining to me what you are referring to? Or am I just being dense?

    Tom, Thanks for confirming my position. I am not sure whether I am in agreement with Nick or not. I haven’t really thought about that particular formulation of his. By the way, my impression is that the required reserve ratio on almost all deposits is way less than 10%. But I am happy to be corrected on that if I am in error.

    As I read through your comment, I think that I may see what Nick is getting at, but maybe not. It seems that he is saying that the central bank sets its lending rate at a particular level for a given period of time. For the duration of that time period, the amount of reserves it creates is demand determined. But the central bank gets to re-evaluate its interest rate target every couple of months or so. If that is what he is getting at, then I would certainly agree.

    nottrampis, I do want to read the paper, but can’t promise when. I will certainly keep in mind the consumer demand for a post on the topic, and as a general proposition I am in favor of responding to consumer demand, within reason of course.


  18. 18 Becky Hargrove February 11, 2013 at 4:39 pm

    Sorry about my lack of clarity. It was not clear where you stood in regard to banking and currency schools, so your answer to Will helped my understanding in that regard, and your answer to Marcus also clarified some concerns with the money multiplier in general.


  19. 19 Tom Brown February 11, 2013 at 8:47 pm

    David, thanks. Now do you agree when Nick writes that over long periods of time in relation to the CB rate adjustment interval that the supply of money is PERFECTLY inelastic wrt the overnight interest rate, but perfectly elastic wrt the inflation rate? The part I have a hard time with is the word “perfectly” in the first part. When envisioning this perfect inelasticity wrt overnight rates, I’m seeing a supply curve, with stock of money on the x-axis, the overnight rate on the y-axis, and a vertical supply curve. Nick states as much here:

    “At the Bank of Canada’s medium term horizon, the supply of money is perfectly interest-inelastic. It’s perfectly income-inelastic. It’s perfectly almost everything-inelastic, except for one thing. It is perfectly inflation-elastic.

    The money supply curve is vertical on the old picture, once we get past 6 weeks. Redraw the picture, delete the rate of interest on the vertical axis, and replace it with the rate of inflation. Draw the money supply curve perfectly elastic at 2% inflation.” — N. Rowe

    The part I don’t see is why this curve is “vertical” before we replace interest with inflation on the y-axis. Why must it be vertical? Why not at some finite non-zero slope?

    Nick has instructed us to mentally draw this vertical supply curve, so I suppose it must convey some useful information, but what I cannot tell. For one thing, at what particular value of the stock of money must it be drawn? If it is at one particular stock, then the locus of points on the curve only allows this one stock as a possibility for the mid-term case. However, the horizontal supply curve (which Nick implies is just another way of looking at the same situation, i.e. the mid-term, or two year case) obviously allows all possible stocks of money at a single inflation rate (the target inflation rate). Thus which is valid? Only one stock of money possible, or all stocks of money possible?

    I’m wondering if it might be more accurate to say that in defending the inflation rate target, the CB does not necessarily control the absolute stock of money. That’s certainly what the horizontal supply curve Nick had us draw implies.

    This leads me to think that if the CB is targeting an inflation rate, the stock of money could vary and the interest rate could vary, but I don’t see where the the stock and inflation rate are correlated with a coefficient of magnitude 1. I would think they would have a smaller magnitude correlation coefficient (i.e. they would have a degree of independence from one another).

    Another way of saying this is that the stock of money is INFLUENCED by the overnight rate and by the demand for borrowing from banks, but neither determines the stock of money completely, even in combination.


  20. 20 Tom Brown February 11, 2013 at 9:05 pm

    David: re: value of required reserve ratio in the US. Don’t take my word for it! 10% is a typical figure I see used for banks in the US in various sources and examples, however I’ve also heard that the the largest banks have negotiated ratios that are considerably smaller than that (i.e. 1/30, 1/50, etc). So what I heard there matches your impression, at least for the large banks. Does anybody actually know what the facts are?

    Wikipedia says this (kind of the opposite of what I’ve written):

    A depository institution’s reserve requirements vary by the dollar amount of net transaction accounts held at that institution. Effective December 29, 2011, institutions with net transactions accounts:

    Of less than $12.4 million have no minimum reserve requirement;
    Between $12.4 million and $79.5 million must have a liquidity ratio of 3%;
    Exceeding $79.5 million must have a liquidity ratio of 10%.[4]


  21. 21 David Glasner February 12, 2013 at 5:07 pm

    Becky, Glad that the picture seems to be getting clearer.

    Tom, Sorry to seem evasive, but I really don’t know if I agree with Nick about whether the supply is perfectly inelastic with respect to the overnight rate and perfectly elastic with respect to the inflation rate. I think he is talking about a market equilibrium relationship between the money supply and the overnight rate and between the money supply and the inflation rate when the central bank is targeting an inflation rate. So the central bank will provide whatever money supply is consistent with the target rate of inflation and will adjust its overnight rate as necessary to achieve the desired inflation rate. My problem with that is that that is not a supply curve in the sense in which we talk about supply curves in analyzing markets. Only one point on an ordinary supply curve is consistent with equilibrium. Every point on Nick’s supply curve is an equilibrium.

    Thanks for the info about reserve requirements. I will have to try to figure out what it means.


  22. 22 Tom Brown February 12, 2013 at 7:13 pm

    David, thanks for your response regarding Nick Rowe’s article.

    Re: reserves, I recall now the source I heard for the statement I made about the largest banks negotiating smaller reserve ratios for themselves: it came from Chris Whalen on an interview he did on the now cancelled program “Capital Account.” You can still find the episodes posted on youtube. But I don’t recall which one it was! I did a little googling to see if I could verify that elsewhere but I couldn’t.

    Another question about something you stated in a recent response to me. You stated that you thought that IOER was a bad policy. I don’t know if it is one way or the other, but I saw an interesting comment on that recently that I wanted to share with you. This comes from “JKH” (the articles author) who’s responding to a comment saying that the whole problem with banks not making loans boils down to IOER (I quote the best part below the link):

    “But here’s a pragmatic question: Excess reserves are now, what: $ 1.5 trillion give or take? Interest of 25 basis points a year on that amount is $ 3.75 billion pretax, which is probably around $ 2.5 billion after tax. Do you really think it’s conceivable that a banking system that makes $ 100 billion plus per year after tax is going to knock the lights out with aggressive new lending just because it loses $ 2.5 billion in reserve interest?”

    What do you make of that response? I think I see why IOER serves as a disincentive to create new loans, but perhaps, as JKH implies here, it’s a very minor disincentive. You’re thoughts?

    JKH goes on to further state, again in criticism of the argument that IOER is making much difference:

    “More fundamentally, market monetarism appears not to understand the difference between bank liquidity management and bank capital management, and the relationship of those two things to bank lending decisions. And in this case it also fails to consider the way in which banks consider overall pricing strategies when faced with any systemic shock to interest margins – they tend to adjust through asset and liability pricing responses that are directional in their effect in preserving return on capital. They don’t try and make it up on volume.”

    I don’t fully understand JKH’s argument here. But what I think he’s getting at is that banks probably wouldn’t try to make up their lost $2.5B by creating… well lets use the 10% required reserve ratio for laughs… by making $15T in extra loans (10*$1.5T in ER). They would instead “adjust through asset and liability princing.” I don’t know what this means, but perhaps it includes lowering the interest they pay on CDs and deposits to their customers (as examples).

    Even if banks went the route of making new loans to make up for lost IOER, they don’t need to convert ALL their ERs to RRs (they don’t need to make $15T in new loans!).. in fact they may only have to marginally increase their lending (probably)… but they’d still be left with a LOT of ERs on their books, don’t you think? So in light of these figures, how much new lending do you think banks might undertake to make up for lost IOER?


  23. 23 stearm February 13, 2013 at 2:01 am

    Let’s think out of the box. Money supply is volatile, prices are sticky and prices of old products have even a tendency to fall. We -households, corporations, banks, governments- all create money when we spend. But firms piles cash, households save, governments reduce deficits, banks receive interests on reserve. Yes, the central banks may even start dropping money from helicopters, but that hardly makes a difference to me and to the public unless someone start to use that money to buy and transform physical wealth into more products and, in particular, new ones. Which is clearly not happening. If I win the lottery tomorrow, I will buy a Picasso. The former owner can now buy a Cezanne and the owner of the Cezanne will buy the first copy of Luther’s Bible. Due to transaction times, that money is under the bed for at least two years (physically it is located somewhere in some bank’s account, but where is the one willing to use the money to pay workers and fixed capital?).


  24. 24 David Glasner February 15, 2013 at 8:40 am

    Tom, The point about IOR is that it means that the banks will always be willing to hold more cash, because holding cash becomes preferable to holding any other asset. If banks cannot be holding too much cash there is no way to induce them to start lending by putting more cash on their balance sheets. It is a true liquidity trap. Without interest on reserves, it is possible to put more money on bank balance sheets than they want to hold, inducing them to trade the cash for something else, which would have some stimulative effect. Quantitatively it might be small as JKH suggests, but with enough effort it cold be done. With interest on reserves, it can’t be done.

    stearm, With enough money in their pockets, people will eventually decide to spend some of it, that is provided that holding money involves some cost (foregone interest). At the zero lower bound, it becomes hard to do this. But there is no sense in making it even harder by paying interest to banks on the reserves they hold.


  25. 25 Tom Brown February 15, 2013 at 6:27 pm

    David, thanks again. BTW, I thought you might be interested, I asked JKH if the Fed pays interest on a bank’s vault cash (paper bills and coins held by the bank and considered to be a form of bank reserves). He claimed it does not. I think I’d asked you once and you were not sure. I don’t know JKH’s credentials, but he does seem to know something about banking, so perhaps he’s correct!

    The reason I asked him was because the same commenter he was responding to above (regarding IOER) went on to ask JKH to consider a hypothetical world where negative IOER was charged (or more generally, negative IOR). The commenter conjectures this would force the banks to lend, but JKH made a good argument, that this would actually have some very disruptive sounding unintended consequences… one of which (I think) is that there’d suddenly become a huge market for paper money, since it would presumably be free of negative IOR (since it’s free of positive IOR now). I actually couldn’t decipher everything he touched on, but here’s the start of the thread:

    The part I’m talking about is where he mentions “bank notes” towards the end, but the rest is interesting too.

    In fact, one of the things JKH brings up makes me think that if IOR were to fall to zero, don’t you think that the banks might start charging interest or larger fees for deposits, since that’s an easy way to keep their current spreads w/o having to actually take on any risky lending activity? What are people going to do? Start holding large amounts of paper currency? I suspect that a lot of us would be willing to pay for the convenience. We already “pay” for the convenience of using credit cards, though indirectly.

    Even if bank assets (ERs in this case) have a NEGATIVE or zero rate of return, the bank still makes money if the rate it must pay for its liabilities is even MORE negative. The old “pass the costs to the consumer” strategy.

    BTW, JKH brings up another one on the long list of ways people refer to paper bills and metal coins: JKH appears to favor “bank notes.” So I’ll update the list:

    1) paper bills and metal coins
    2) currency
    3) cash
    4) bank notes
    5) dead presidents

    The way you use “cash” in your latest response to me implies you do NOT mean paper bills and metal coins.

    I tend to agree with you that having IOR <= 0 makes banks eager to get rid of their ERs, but making loans to non-banks won't accomplish that since it only converts a small percentage of the loan principal amount in ER to RR (and presumably by IOR you meant that to applies to ALL bank held reserves — except vault cash, correct?). Of course I'm speaking in aggregate again, and at the level of an individual bank, since almost all loans are made with the idea of purchasing something, then there is a good chance that the lending bank can send along it's negative IOR reserves to the lucky receiving bank to clear the payment associated with the reason the loan was taken in the first place. Of course if the seller happened to have an account at the same bank, the loan does them no good in this regard. And we already covered the ways that reserves can leave the banking system (actually, JKH took my list and expanded it and made it more exact in that same post). So on aggregate, it doesn't matter: banks are stuck with the reserves until they can get rid of them (mostly) with the aid of the government or the Fed. But as seem from the perspective of individual banks, they can do all they can to keep expensive reserves OFF their balance sheets in favor of other assets. But there will be winners and losers to that game, but perhaps it makes the velocity of money increase? Is that what you're getting at?

    BTW, JKH's more complete list of the only ways in which reserves leave (or enter) the banking system (in aggregate) is here (he brings up Bernanke's salary for one!):


  26. 26 David Glasner February 21, 2013 at 12:26 pm

    Tom, I’m not sure why stopping paying interest on reserves would be disruptive. There are substantial benefits from holding deposits rather than cash for banks as there are for individuals. I also believe that the Swedish central bank is charging negative interest on reserves held at the bank. So it’s a question of setting the interest rate at the right level. But first let’s stop paying positive interest on reserves.

    If banks were not receiving interest on reserves they might raise fees on bank deposits might start charging higher fees and pay less interest on deposits, but from a macroeconomic perspective that would not have any bad consequences unless people switched to currency, and the Fed could counter that by issuing additional currency. But the main point of stopping interest on reserves is to make it easier to raise the price level and increase inflation expectations which would tend to raise nominal interest rates and move away from the zero lower bound.


  27. 27 Tom Brown February 21, 2013 at 5:22 pm

    Do you think the Banks would lobby hard to prevent IOR from going away? What are the political prospects for IOR actually going away? Banks tend to get whatever they lobby for.


  28. 28 Will February 24, 2013 at 2:25 pm

    Thanks, David. I’m going to put your book on my wish list for my upcoming birthday. I clearly need to read it. Just out of curiosity, am I wrong in seeing Hayek, Mises, and Austrians generally as currency school types? If so, the teacher who explained the two schools to me has some explaining to do.


  29. 29 David Glasner February 25, 2013 at 9:37 am

    Tom, I have no doubt that banks would not be pleased if IOR were stopped, and you are right that they have a way of getting what they want out of the political process (including the Fed). Beyond that, I don’t have any estimate of the odds that IOR will be terminated.

    Will, Thanks, you made my day. About Hayek, Mises, and the Austrians, you are right that they were totally in the Currency School camp. I once had a conversation with Hayek after he began writing about competing currencies and the denationalization of money, which is some ways marked a change from his earlier monetary theory, in which I tried to suggest that he had misunderstood that Banking School position. He listened politely, and said that he had not looked at the Banking School literature in a very long time, and conceded that it was possible that he had not interpreted them correctly, but that was as far as I could get with him. Other Austrians are generally much less open to any concession of error or misunderstanding.


  1. 1 Friday, February 8 – Daily Reading List Trackback on February 8, 2013 at 6:46 am

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

About Me

David Glasner
Washington, DC

I am an economist in the Washington DC area. My research and writing has been mostly on monetary economics and policy and the history of economics. In my book Free Banking and Monetary Reform, I argued for a non-Monetarist non-Keynesian approach to monetary policy, based on a theory of a competitive supply of money. Over the years, I have become increasingly impressed by the similarities between my approach and that of R. G. Hawtrey and hope to bring Hawtrey’s unduly neglected contributions to the attention of a wider audience.

My new book Studies in the History of Monetary Theory: Controversies and Clarifications has been published by Palgrave Macmillan

Follow me on Twitter @david_glasner


Enter your email address to follow this blog and receive notifications of new posts by email.

Join 3,263 other subscribers
Follow Uneasy Money on

%d bloggers like this: