Banks and the Monetary Base – A Friendly Response to Paul Krugman

Paul Krugman is unhappy with me nipping at his heels about banking.  But I think this is important and crucial to Dr. Krugman’s idea that the “synthesis” has been “lost” so I am taking hold of that ankle once again and shaking my head as vigorously as I can.    Anyhow, he says his model hasn’t been wrong and that he’s just oversimplifying the way banking works while remaining totally consistent with Tobin-Brainard &  Diamond-Dybvig.  I don’t see it that way.  Let me explain.

1)  First of all, I think Dr. Krugman might have me confused with some other people who think they’ve “discovered the hidden secrets of the monetary universe”.  Lots of people confuse MMT for being the same as Post-Keynesian Economics, but that’s not quite right.

Much more importantly, I think it’s crucial to embrace the fact that “generations of economists” probably have misunderstood how the monetary machine works to some degree or at least don’t have views that perfectly apply to today’s system.  After all, the monetary system is not a stagnant system.  It is dynamic and evolving.  The system in place in 1900 is VERY different than the system we have today.  That requires constant upkeep and tweaking of core concepts and models.  Not to mention, economics as a whole is a rather new development. I think it would be silly to assume that we’ve reached “peak economics” in terms of our understandings when economics as a discipline has only existed for a few generations.  We’re very much in the early stages of developing our understandings of these things despite the many great minds that have researched this subject matter.

2)  The two papers Dr. Krugman cites were written in 1963 and 1983 long before QE with IOR was implemented and during a time when policy was in fact very different.  The way monetary policy is enacted today is dramatically different than it was in 1963.  More importantly, the way policy impacts banking is dramatically different than it was in 1963.  For instance, the Tobin paper discusses in detail how reserve requirements are used as a policy tool to control reserve balances and influence the way banks lend.  But that view is entirely inapplicable to a world where the Fed is paying interest on reserves when their balance sheet has been expanded.  In other words, the cost of funding (what is traditionally thought of as the Fed Funds Rate) has become the Interest on Reserves Rate.  I.e., the IOR is a de facto FFR.  As point 1 expresses, you must account for substantial changes in the way the system is designed before you conclude that someone’s 50 year old views are a good stepping stone for understanding the current system and current policies.  I am fairly certain that James Tobin would agree that paying IOR changes the way we should view the system.

3)  What I was really saying in my initial post was basically an expansion of this thinking – that in order to understand why QE didn’t cause high inflation you need to view the monetary system differently.  And the IS/LM model wasn’t going to get you to the right conclusion because it doesn’t actually apply (rather, it might get you there, but not for the right reasons).  Stick with me here if you’re a non-economist.  This will be a little wonky, but you’ll get it.

If you look at the diagrams below we’ll basically see the model of the world as Dr. Krugman views it at present where we’re in the “liquidity trap”.  Normally, when the Fed increases the money supply (pushes money supply 1 to money supply 2 – the blue bars) the interest rate goes down.  Normally, if we reduce the real interest rate then investment/saving increases as GDP increases (see figure 2).  But when you’re at the zero lower bound where interest rates can’t go down any further then the Fed hits a wall and “printing” more money (going to Ms3) won’t increase GDP.

LiquidityTrap (1)

There’s a few problems with using this model.  First of all, the Fed isn’t determining the interest rate by altering the money supply.  The reason they’re paying interest on reserves is specifically so that they can control interest rates without having to worry about the amount of reserves in the system.  In other words, unlike the James Tobin world, interest on reserves is a de facto Fed Funds Rate.  So the quantity of money in the reserve system is not the determining factor of the overnight interest rate.  If the Fed wants to raise interest rates it will not look at the chart above and shift Ms3 back to Ms1.  IT DOESN’T HAVE TO!  The Fed will simply raise the interest rate on reserves which will stop banks from trying to lend reserves to one another (because lending reserves to one another – the only time they actually do lend reserves – would drive the rate down).

In other words, the charts above, which are quantity centric views of the monetary system don’t apply with IOR because the Fed isn’t targeting quantity, they’re targeting price!  So the charts above tells us a lot less than Dr. Krugman’s model claims because the quantity of money isn’t the determining factor of the interest rate.*

Second, the main point of my previous post was to highlight the fact that banks don’t make lending decisions based on the amount of reserves they hold.  And yes, James Tobin most certainly understood this point.  In  “Commercial Banks as Creators of Money” he wrote:

“An individual bank is not constrained by any fixed quantum of reserves. It can obtain additional reserves to meet requirements by borrowing from the Federal Reserve, by buying ‘Federal Funds’ from other banks, or by selling or ‘running off’ short-term securities.”

In other words, banks lend first and find reserves LATER if they must.  This is in stark contrast to what Paul Krugman writes 50 years later:

“First of all, any individual bank does, in fact, have to lend out the money it receives in deposits. Bank loan officers can’t just issue checks out of thin air; like employees of any financial intermediary, they must buy assets with funds they have on hand. “

Banks are the primary creators of money in our monetary system and as Tobin correctly notes, they can lend first and find reserves after the fact.  But this starts on the demand side, not the supply side.  In other words, banks lend when creditworthy customers have demand for loans and not just when creditworthy banks have access to supply.

Now, it’s important to understand that most of what the Fed actually does is just asset swaps to help try to hit an interest rate target or try to influence the private sector portfolio holdings (QE is not actual “money printing” in any real meaningful sense because it’s just an asset swap).  This is why it’s crucial to understand precisely what QE does.  QE swaps one privately held asset for another.  When QE is performed with a bank the bank ends up holding more reserves and fewer t-bonds. In the old textbook model the government has increased the “money supply”, interest rates will fall and banks will lend more as investment increases and banks “multiply” their reserves.  But reserves are held in the interbank market and don’t influence the amount of lending banks make (per Tobin).  So increasing the level of reserves is not analogous to firing dollar bills out the front of the Federal Reserve.  And contrary to popular mythology, the Fed is not the actual distributor of cash notes so that’s a diversion in all of this.  The quantity of cash notes held by the public is determined by depositor demand and the US Treasury ultimately will fill orders from regional Fed banks to meet bank customer demand.  The Fed doesn’t push cash out of its doors.

So, QE and more reserves didn’t cause rates to go down because we know the Fed can control rates without having to worry about the quantity of reserves.  And we know that the quantity of reserves didn’t cause the money supply to increase in any meaningful sense because it doesn’t result in more lending and amounts to little more than an asset swap.  So the IS/LM model, which is quantity theory based, didn’t tell us as much as one might presume.  So there had to be something else going on there.  Of course, I’ve argued that the demand for credit was low because households were de-leveraging.   And that would explain why borrowing has been stagnant and economic growth has been rather anemic. In other words, the REAL transmission mechanism to increase private sector money (inside money or bank money) was broken!  And we didn’t have to understand the IS/LM model to understand why QE wouldn’t cause high inflation.

There’s a more interesting discussion to be had here and one I hope Dr. Krugman will entertain.  But I will discuss this in detail tomorrow because I fear if you’ve made it this far your eyes are starting to get awfully heavy and my ankle biting is losing its efficacy…..

* We can quibble over QE’s impact on long rates, but I’ll take the SF Fed’s word for it that it’s “negligible”.  

 

-------------------------------------------------------------------------------------------------------------------

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

Cullen Roche

Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services. He is also the author of Pragmatic Capitalism: What Every Investor Needs to Understand About Money and Finance and Understanding the Modern Monetary System.

More Posts - Website

Follow Me:
TwitterLinkedIn

Comments

  1. Cullen, in the above, you have “which would drive the rate down.” in the part right under figure 2. Should that read “up” rather than “down?”

  2. Great! Looking forward to tomorrow’s update! (my eyes aren’t heavy at all… don’t know about Krugman’s ankles though) :D

  3. I think he means that the Fed is setting IOR in order to deter banks from trying to lend their reserves to one another because that would put downward pressure on the overnight rate.

  4. Also, why are figures 1 & 2 entitled “Liquidity Trap?” You text sounds like they describe something which is NOT the liquidity trap (ZLB) but rather how it would work if we were not at the ZLB.

  5. DanH: so you’re saying I’m not parsing that correctly. It should be parsed like this instead:

    “It [the Fed] will simply raise the interest rate on reserves which will stop banks from trying to lend reserves to one another [because lending reserves to one another] would drive the rate down.”

  6. You’re the man Tom.

    Edited:

    The Fed will simply raise the interest rate on reserves which will stop banks from trying to lend reserves to one another (because lending reserves to one another – the only time they actually do lend reserves!) would drive the rate down.

  7. He says it “makes his flesh creep” just to type that! LoL!

    BTW, if Nick is correct, and supply = a curve while stock = single value, then what is the corresponding single value word in relation to “demand” if demand is also a curve?

  8. Well, that sure is succinct considering the breadth of subjects touched. Unfortunately, my head is spinning a little bit.

  9. Cullen, one thing though, regarding the causality between loans and deposits and Krugman’s (and Sumner’s) assertion that it really runs both ways and is thus a “simultaneous system.” I’m not sure I see you addressing the logic (or mis-logic) of that statement here. How do these guys (Krugman and Sumner) come to this conclusion? What’s the justification? Krugman cites Tobin on that (links in blow):

    http://ask-cullen.com/krugman-cites-tobin/

  10. It doesn’t run both ways. What Krugman is confusing is that banks are capital constrained and not reserve constrained. The quote Cullen used from Krugman makes this crystal clear.

  11. I am specifically referring to the lending function. Banks don’t lend their money as Krugman has stated. As Tobin said, they can lend and borrow from the Fed if they must. IF they must…..

  12. I agree that it doesn’t run both ways. I don’t see it both ways. I see loans create deposits. I’ll take another look at Krugman and see if I agree that that’s how he came to that conclusion…

    The part I’m specifically referring to in Krugman’s piece is this (especially the 2nd paragraph):

    “Actually, Tobin-Brainard is to many of the controversies that swirl around banks and money as IS-LM is to controversies about interest-rate determination. When we ask, “Are interest rates determined by the supply and demand of loanable funds, or are they determined by the tradeoff between liquidity and return?”, the correct answer is “Yes” — it’s a simultaneous system.

    Similarly, if we ask, “Is the volume of bank lending determined by the amount the public chooses to deposit in banks, or is the amount deposited in banks determined by the amount banks choose to lend?”, the answer is once again “Yes”; financial prices adjust to make those choices consistent.”

    Sumner liked that so much he says he’s going to adopt “It’s a simultaneous system” as his pat answer to people bringing up this causality issue. I haven’t read Tobin… I’m just wondering if there’s something in there that gives them this justification.

  13. OK, I got that. I’m just still confused about how Krugman/Sumner supposedly see causality running both directions. Krugman is saying (in the quote you give in the post) that deposits come in and banks lend them out. From the Krugman quote I give above, he’s saying both of these things:

    1. “bank lending [is] determined by the amount the public chooses to deposit in banks”

    OK, I see why he’d say that and I understand why it’s wrong, but he also says this:

    2. “the amount deposited in banks [is] determined by the amount banks choose to lend”

    How can he say 2 (which is correct) but also say 1 (which is not) and claim they’re the same thing?… part of a “simultaneous system.” I don’t see the world view that allows both.

  14. I don’t see anything even remotely controversial in this piece. Heck, even Johnny likes it! Mr. Krugman is going to have a tough time poking holes in it.

  15. Hope Krugman is game. Very well written, can’t wait to read the next post. Nice one Cullen.

  16. Cullen, still a couple of things about this sentence and the accompanying figures 1 and 2:

    ” If the Fed wants to raise interest rates it will not look at the chart above and shift Ms2 back to Ms1. IT DOESN’T HAVE TO! The Fed will simply raise the interest rate on reserves which will stop banks from trying to lend reserves to one another (because lending reserves to one another – the only time they actually do lend reserves would drive the rate down).”

    1. The easy one (nit): I think you should write that last sentence like this “(because lending reserves to one another – the only time they actually do lend reserves — would drive the rate down)” … in other words it’s *slightly* more clear with that 2nd “–” in there.

    2. Regarding raising interest rates and not “having to” shift Ms2 back to Ms1: Isn’t the real problem here NOT that we’re at Ms2, but that we’re at Ms3, which is significantly to the right of Ms2 (i.e. Ms2 << Ms3)… but that rates have flattened out: i.e. not decreased lower than your horizontal red "Liquidity Trap" line (in figure 1). So now to raise rates again, we've first got to get rid off all the excess reserves which represents the difference between Ms3 and Ms2 (ER = Ms3 – Ms2). So rather than "having to" shift from Ms2 to Ms1, the Fed would prefer to use that mechanism to raise rates, but since we're not even at Ms2 yet, they HAVE to resort to raising IOR.

    3. If I'm correct about Ms3, interest rates, and your red line in figure 1, then this brings up an interesting question about figure 2 which I'm having a problem with: Suppose we're actually at Ms3 WELL to the right of Ms2 in figure 2. Does the liquidity trap line really describe the situation though?? Sure the interest rates can't go lower (vertical axis), but GDP does NOT really go up as M is moved to the Ms3 position or even further to the right, correct? GDP, seems like it too is blocked by the "Liquidity Trap" in some regard.

    So, forget about 1. (that's a stylistic nit), but if I'm correct in interpreting 2, then how do you respond to my point 3? Shouldn't that diagram be altered in some way? Perhaps get rid of the vertical Ms1 and Ms2 bars and instead just show the points on the curve where the interest rate corresponds to Ms1 and Ms2 (i.e. they might be better as horizontal bars in Figure 2). But in figure 2, where would Ms3 lie? I say it would lie right on top of Ms2.

    Or perhaps the labels on the axis could change: you could have the x-axis in figure 2 be "Q" instead and then place "GDP on the y-axis: then the Ms1, Ms2, and Ms3 lines would be as I described for figure 1 (vertical, but arranged with w/ Ms3 to the right of Ms2 and Ms1 as you currently have them). But you could make the black curve rising instead of downward sloping and put your horizontal red "Liquidity Trap" line up where Ms2 meets the curve… thus showing how GDP is limited by the liquidity trap despite a growth in Q from Ms2 to Ms3.

    Make sense?

  17. Keep going Tom!

    In answer to your question: there are three scalar quantities:
    1. quantity demanded (that’s the point on the demand curve)
    2. quantity supplied (that’s the point on the supply curve)
    3. actual quantity.

    Only in equilibrium are all three quantities the same.

    If we are out of equilibrium, ***for any good except money***, a sensible assumption is that actual quantity traded will be whichever is less: quantity demanded or quantity supplied. So that if the price of apples is above equilibrium, the actual quantity traded will equal quantity demanded, which is less than quantity supplied. And if it’s below equilibrium, actual quantity traded will equal quantity supplied, which is less than quantity demanded.

    But money is not like other goods. That’s because while apples are only traded for money (in a monetary exchange economy), money is traded for all other goods, in all markets. There is no “money market”. All markets are money markets. And so sometimes people want to “buy” more money, not because they want to *hold* more money (which is what “quantity of money demanded” means) but because they want to turn around and “sell” it again. And what this means is that central banks can increase the quantity of money above the quantity demanded. Money is really weird like that. The simple picture of a demand and supply curve in “the market for money”, with the central bank having a perfectly elastic supply curve setting “the price of money” (an interest rate isn’t the price of money because all prices are the (reciprocal of) the price of money) just doesn’t work.

    (Minor point: a curve only has two dimensions, and we really need to talk about supply and demand functions, not curves. And modern central banks target inflation, not the rate of interest, so the supply of money is perfectly elastic with respect to the rate of inflation, not the rate of interest.)

  18. Nick! … Thanks for piping up here!!… I read that pretty quick, so I need to go back and review it again, but thanks for the clarification.

    but just looking at the last paragraph (what I can see while typing this) you say:

    ” a curve only has two dimensions”

    Yes, a curve can be represented on a 2-D plane, or a 3-D plant, or an N-D plane … or even a 1-D “plane,” but I’d argue that it really just has one dimension: namely how far along the curve you are, which can be represented with a single scaler. A scaler is essentially a 1-D “surface” … and a “surface” we normally think of occupying 2-Dimensions (needed at least 2 degrees of freedom: i.e. two scalers to determine where you are on it). Of course I think you could generalize and talk of a 3-D surface occupying 4 or more dimensions, etc. At least that’s what I learned in math class! ;)

    I love the fact that you brought up “elastic” … that’s one of my favorites from you (it made me go on a quest one time to figure out what the heck you meant by that).

    OK, I’ll re-read and try to absorb it all this time.

  19. The extraordinary measures during the GFC showed (to monetarists) that there must be more to it than base money (you wud of thunk!). Why do you think there was emergency asset purchases, securities lending, guarantees on money markets, intervention in trade financing, FX swaps, bailout of the banking and insurance system, …. !?!?

  20. Cullen, I’m still confused by your new charts with Ms3. The liquidity trap line doesn’t do anything to prevent movement along the curve (in either figure) from Ms2 to Ms3.

    My figure 2 had GDP on the y-axis… I think if you want Int rates on y and GDP on x, then M2 and M3 are the same point aren’t they? (we’re not plotting Q in this case anywhere… but M1, M2 and M3 all correspond to different points on the curve in figure 2).

  21. BTW, when someone posts a link to your “The Supply of Money is Demand Determined” post, to you get a Twitter alert or email or something? That was really funny! That’s one of my go-to Nick Rowe pieces… even though I think you told me one time that it’s probably one of your most “heterodox” in the sense that what you write there is not accepted very widely…. or am I getting that confused with the post you did right before that one on hot potatoes?

    http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/03/banking-mysticism-and-the-hot-potato.html

    I refer to both of those over and over again… not that I agree entirely with either, but they lie at the heart of what I have a hard time with regarding MMist ideas.

  22. Cullen, it all looks pretty good (as far as I can tell!)… but perhaps you should at least mention that in moving from Ms1 to Ms2 (or back again) you are in Tobin’s world… it’s only when we push on to Ms3 that we leave that world. I think you removed all references to Ms2.

  23. This is a great post. Now for the most important part – how do we ensure that Krugman reads it and responds???????

  24. Well Nick Rowe read it, and Krugman reads Nick Rowe (I think) and I think it’s clear he reads Cullen too (especially when his name’s in the title). So if Krugman doesn’t respond, hopefully Rowe will put up a post telling Cullen where he’s wrong (and/or right) and then Krugman will read that! (What may be a lot more interesting though is what Rowe thinks!).

    The other thing you can do is if Krugman posts something even vaguely touching on a similar topic, and you think this post speaks to that … simply post a link back to it on the NYT comments (it’s free). Krugman (as far as I know) never responds in the comments section, but I think he does look them over (once in a great while he talks about a comment he got).

  25. When a central bank pays IOR that will be the floor for interest rates in the economy, since a bank with excess reserves will need a higher interst rate from another bank than from the CB. This is because of default risk.

    So do not understand what you mean by not lending reserves to one another. If there are excess reserves, and one bank is short, it got to lend from other banks, or from the CB if other banks do not want to lend to it.

    Guess we talk about different things, because I agree with your post. What I got anyway.

  26. I think he’s saying that the banks won’t lend at lower than the floor rate. That’s all.

  27. Tom: I have spies all over the internet!

    Nearly all “orthodox” New Keynesian economists, and many market monetarists, will think I am wrong in both those posts.

    But I must go and argue with Cullen.

  28. “That was really funny!” … namely because that’s NOT the first time that’s happened! You showed up another time on another blog shortly after I posted that link. Ha!

  29. Is it just me or are we seeing the walls fall around New Keynesian economics as we speak?

  30. Cullen: The Bank of Canada has been paying interest on reserves for many years, so we have gotten used to it up here.

    Let’s go the whole hog, and assume the central bank pays interest on currency, as well as on reserves. It pays the same rate of interest on all its monetary liabilities. How does that change things, in the ISLM model, for example?

    It means the central bank now has two ways to shift the LM curve:

    1. It can shift the LM curve by shifting the money supply ***curve*** (yes, we do need to be especially picky on terminology now!) Just like before.

    2. It can shift the LM curve by changing the rate of interest it pays to people who hold central bank money, which will shift the money ***demand curve*** (if we draw that curve as a function of rates of interest on other assets, or income, or the price level, like we normally do). That’s new.

    But even that second way isn’t really new. We are familiar with using the ISLM model when the expected rate of inflation changes, so there’s a change in the gap between real and nominal interest rates. (We either shift the LM curve, or stick a wedge between the IS and LM curves.) And inflation is just like paying a negative real rate of interest on currency.

  31. Nick, is this what you meant when you wrote

    “But I must go and argue with Cullen”

    Whew! Good… I thought you’d moved this to Twitter or something (I’d go look if I had too, but I don’t normally like going there).

    Well, I don’t understand this. You officially exceeded my limited grasp of this stuff… so I hope Cullen explains it… ***with charts!***

    … or maybe you’d prefer to do that Nick? You love drawing pictures with words, but some of us have a hard time following those “mental graphs.”

  32. Nick, I just now noticed this brand new post from you touching on some of these issues:

    http://worthwhile.typepad.com/worthwhile_canadian_initi/2013/08/is-money-exogenous-or-endogenous.html#more

    … I’ll dig in!

    I made my own amateurish post on this a while back, with a lot of unnecessary words… not so much an explanation, but my view of this related to the engineering concept of a feedback control system:

    http://brown-blog-5.blogspot.com/2013/06/inflation-targeting-as-feedback-control.html

    (probably just the two figures *might* be of interest to you… especially figure 2… I’m not presenting anything new… but once we translate into a discrete time feedback control system with sample rates and some sort of model for the “plant” etc, then there’s literally entire libraries of well established knowledge we can bring to bear on this from the engineering literature… for example, is the system even “controllable” or “observable” given the current “measurements” available and system states, and control inputs, etc. What about noise… white or colored? Uncertainties in the underlying model? Linear vs non-linear plants and controllers. Time invariance: do we have it? Are there uncontrollable or unobservable states that will cause instability or frustrate our attempts at stabilizing it or controlling it, and what is the system’s time constant, phase margin and gain margin? etc etc etc: I realize that’s all a little far fetched given our understanding of the plant (system to be controlled) at this point… but still, all those tools exist and are really the easy part!)

  33. * Had to edit this comment. I was thinking about coins in my last response for some reason.

    Okay, but that’s a lot like fiscal policy, isn’t it? Now you’re having the Fed pay interest on all outstanding cash which means that the Fed is now acting a lot like the US Tsy does when it taxes people by redistributing money from some people to others except in this case the Fed is earning interest from its operations and paying expenses out of this to meet its cash interest liability….

    We’re starting to blur the lines on monetary and fiscal here, no? I’ve never thought about this so I might not be thinking of it correctly….

  34. Plus, operationally, isn’t this completely unrealistic? How would you even implement this sort of policy of paying interest on cash? How would you track who earns what and when?

  35. You’re going to have a great career as a public intellectual, if that’s what you want. You are going to have an impact on policy, as a commentator, at least and maybe more.
    I can easily see the WaPost or the NYTimes asking you to write for them. I can see a presidential candidate putting you on his staff to advise him. You can explain economics, you are working hard, you realize there is more to money than just money. You’re willing to engage your readers.
    It’s kind of cool watching it happen.

  36. Deficit spending is not redistributing inside money. Unless your claim is that your money disappears upon being deposited into your securities account at the fed.

    Alternatively, you could say that deficit spending is redistributing inside money, but when the Fed redeems the securities, then that is brand new money (deposits or reserves depending on the buyer of course). Because Congress can’t pay to redeem securities out of the Treasury General Fund even if they wanted to. Only the Fed can redeem securities, as evidenced by the fact that the Clinton surplus required zero TGA $’s to be appropriated by Congress for principle repayment

  37. Auburn, I’m not sure I follow you completely, but Cullen referred to “taxing people” as redistributing in his comment above. However, you’d be right to claim that he’s said the same regarding Tsy deficit spending. I think I’ve done an OK job of capturing what he means wrt deficit spending here (esp. on BS sets 2, 3, and 4):

    http://brown-blog-5.blogspot.com/2013/05/banking-example-8-treasury-deficit.html

    “Taxing” would follow a similar pattern. Any objection that that?

    BTW, what’s the Treasury General Fund? Is that the same as the TGA?

  38. You’re using the MMT framework. I don’t use that framework. I use a flow of funds framework. So deficit spending takes inside money from a bond buyer, redistributes it to the recipient of govt spending and the bond buyer receives a t-bond. The monetary system is built around pvt banks and inside money so in order to properly understand it you have to start with banks and add the govt on second. You can’t start with the govt or you imply that we don’t have a bank centric system. I don’t agree with that presentation.

  39. Cullen: well, as soon as the government owns the central bank, and the central bank gives all its profits to the government, (i.e. for all modern central banks, AFAIK), we can’t avoid blurring the lines between monetary and fiscal policy. Because almost any change in monetary policy will change the profits of the central bank which will change the government’s revenues. (The inflation tax associated with increasing the growth rate of the money supply is the classic example of this.) But, as long as we are not talking about very big monetary changes, the fiscal implications aren’t very big. (The Bank of Canada’s profits are normally around 0.2% of GDP, IIRC, and those sorts of numbers were swamped by the size of fiscal changes in the recent recession.)

    Provided we are on the good side of the Laffer curve, an increase in the target rate of inflation will increase the central bank’s profits from the inflation tax, and an increase in the rate of interest the central bank pays on base money will reduce the central bank’s profits. The two are basically equivalent (except for opposite signs). A 1% increase in the inflation target coupled with paying 1% higher rate of interest on base money should have a roughly zero effect on the central bank’s profits and the government’s fiscal position. People lose an extra 1% through inflation depreciating the base money they hold, but gain that same 1% back through earning higher interest.

    Another name for the fiscal effect of monetary policy changes is the Pigou effect (or money-wealth effect). But since base money is normally only around 5% of annual GDP in advanced countries (maybe double that in the US, because lots of foreigners hold US dollars too), those fiscal effects aren’t normally very big (unless we are talking Zimbabwean levels of money printing).

    Nearly all the action from monetary policy comes from substitution effects rather than from those income or wealth effects that are the basis of normal fiscal policy like tax cuts or increases.

  40. Operationally, yes, a bit hard to implement. One way would be to hold a lottery, and if the $20 note you are holding has the lucky serial number, you win a big prize. The Brits once did this with “Premium bonds” (as alluded to in Jethro Tull’s classic song “Thick as a Brick” in the line “Good old Ernie, he’s coughed up a tenner in a Premium Bond win” [Ernie was the random number generator] sorry for the prog-rock flashback.)

    But there’s no operational problem with doing it with commercial banks’ reserves held electronically at the central bank, of course. So part of the base already pays interest, but not all of it does. Rather than arguing with you by saying “but currency doesn’t pay interest!” it seemed more useful, as well as simpler, but follow you to your argument’s logic, in a world where all base money pays interest.

    And who knows what the future will bring. There’s no reason why ordinary people, and not just commercial banks, couldn’t hold central bank money in electronic form rather than paper.

  41. Hey Nick,

    I like that way of thinking about these things. Too often we try to draw a line in the sand declaring what monetary and fiscal policy are when there’s quite a bit of overlap in many cases.

    Not sure if you saw this post by me, but i’ve done some thinking on how to converge fiscal and monetary:

    http://pragcap.com/the-monefiscalist-synthesis

    I think David B is pretty open to this sort of thinking. Scott S seems pretty averse to it. I don’t think he likes me very much! Oops. I didn’t ever intend to be rude to him, but I think he took some of my comments the wrong way. Oh well.

    I am a post-kesynesian influenced thinker, but not one of these PKE guys who completely shuns monetary policy. I think an endogenous money thinker who does that is being pretty closed minded.

    Anyhow, hopefully there’s overlap in our work.

    Thanks for commenting. Good stuff.

    Cullen

  42. FWIW, it would be a logistical nightmare and an unwise idea IMHO. But I’ll take a crack as a thought experiment:

    Perhaps the Federal Reserve could impose a negative interest rate (or an interest charge) of 0.25% on all demand deposits held at member banks. Thus, relative to same amount in demand deposits, a given quantity of cash would appreciate in value.

    Of course, this will never happen.

  43. right Tom, typo on my part. TGF should read TGA.
    On the taxing front, I would say that you present an accurate explanation. Because you and I don’t get our dollars back from taxation, those are gone to someone else forever. In that sense, taxation is always a pure redistribution of bank deposits.
    However, when I buy a T bond and deposit my money at the Fed for a time, Congress then has authority to spend that amount. Now, we can debate whether or not this is my deposit money that is literally given to someone else (my claim would be that its not or at the least that its irrelevant). But when my T-bond matures, unlike taxation, I get all my money back.
    One part of that flow has to be new money. The money is either new when the Govt spends it or its new when the fed lets me withdraw all my deposit at maturity.
    As I wrote above, it would be one thing if congress had to use surplus dollars to literally pay back maturing securities. Then I think we could fairly describe this part of the process redistributing bank deposits and not creating any new money. Put as the Clinton surplus era proved, this is not the case. There is no such thing as Congress appropriating funds from the TGA to pay back maturing T-bonds. That is the Fed’s responsiblity.
    Anyways, thats my position. I feel it makes sense given my current knowledge of reality about this particular matter.
    For the record, for all intents and purposes, your T accounts looked good to me, accounting layman that I am.

  44. we fundamentally disagree on whether the current Govt, as it is constituted at present is capable of creating new money, bank deposits specifically. And thats good, disagreement is healthy.

  45. Auburn, you wrote:

    “when I buy a T bond and deposit my money at the Fed for a time, Congress then has authority to spend that amount. ”

    Did you mean to write “deposit my money at the Tsy for a time” … because that’s what happens when you get on Tsy Direct and buy a bond from Tsy at auction (of course this involves an bank intermediary, as always). Because buying a bond from the Fed just means your is debited by the intermediary bank, and the bank’s Fed deposit is likewise debited by the Fed (thus both inside and outside money cease to exit).

    When Cullen speaks of Tsy deficit spending resulting in a redistribution it’s exactly as I’ve presented it in those example balance sheets: Person y’s deposit is debited by Bank A, Bank A’s Fed deposit is debited by the Fed AND Tsy’s TGA is credited by the Fed, so then when Tsy spends the proceeds (in my example, on Person x) this process is reversed, except to x and y’s deposit has thus been redistributed.

    I agree that the difference between this and taxing is that a brand new T-bond has also come into the hands of Person y (i.e. the private sector also picks up a new net financial asset (NFA) in addition to the redistribution that takes place). And if Tsy finances that maturing NFA by more deficit spending, then that just means that the old NFA turns to bank deposits but a new NFA is issued… again the net effect is nothing (a deposit was lost by the public to buy the new bond, the deposit is redistributed to the old bond holder (and his bond ceases to exist) and a new NFA is issued.) I illustrate something very similar here (but with the Fed involved buying the boinds and holding them to maturity):

    http://brown-blog-5.blogspot.com/2013/03/banking-example-6-cb-holds-gov-debt-to.html

    What matters in the end for the public’s accumulating equity (in this simple model) is the negative equity of the Tsy… rolling over bonds does not increase this negative equity. I illustrate that point here:

    http://brown-blog-5.blogspot.com/2013/08/banking-example-11-all-possible-balance.html

    If you look at the simplest two sets of BSs at the very bottom, it’s clear that in this four entity would the public’s equity will be $T which is the same as the Tsy’s negative equity (i.e. the debt). Of course that’s a huge simplification and ignores all the good things that come from private investment and real assets, etc… but just on this very simple financial way of looking at it, that’s the case.

    In the above link I provide an interactive spreadsheet and balance sheets to play with to illustrate this further.

  46. Auburn, I posted this same link above, but I do so again, because I think it addresses your question here:

    http://brown-blog-5.blogspot.com/2013/08/banking-example-11-all-possible-balance.html

    The bottom simplified balance sheet(s) for the public show that the non-bank private sector’s (public’s) money (bank deposits + cash) is:

    public’s money stock = L + B + F

    L = bank loans
    B = bank held Tsy debt
    F = Fed held Tsy debt

    So the public’s money is not directly dependent on T (T = total Tsy debt). The public’s EQUITY is, but not the money.

    I’m igorning GSEs, the foreign sector, and MBS here, but in this simple model, that’s what we’ve got (I’m not ignoring cash).

  47. I think so as well. Why would a bank lend reserves to another bank below the floor when they can just hold interest paying reserves? Furthermore, as long as banks have excess reserves, being short of reserves shouldn’t be a problem.

    Does IOER also present a hurdle rate to making a new loans to non-banks due to the risk that the bank deposits could move to another bank?

    If a bank makes a loan at a rate equal to the IOER, and the deposit created leaves the bank, wouldn’t this result in the bank’s excess reserve balance being reduced?

    I highly highly doubt a .25 IOER is restricting lending, but I could see a situation where the Fed increases the IOER to a high enough level to disincentivize lending at a rate below the IOER rate.

  48. I believe this line is your problem. And it is also my point that you have twice not answered.
    “And if Tsy finances that maturing NFA by more deficit spending, then that just means that the old NFA turns to bank deposits but a new NFA is issued”
    Yes, that is how it works when there is continued deficit spending, rolling over the debt as it were.

    But the case is made clearer when considering the example of a budget surplus environment. Lets say that the Federal Govt ran a budget surplus of $1 dollar ever year for 30 years. And the Govt maintained exactly its current rules with regards to securities issuance. Almost every outstanding bond would have matured. That exact amount of money would have been “paid back” by the Fed to the bond holders as either reserves or bank deposits. And yet, Congress would never have had to acquire $16T in deposits from the private sector to redistribute to the bond holders.
    This is the part you are not addressing.

  49. Thanks Cullen.

    Sort of related to your post on fiscal/monetary: (I know you’re not MMT, but that only increases the chances you will get this point).

    Suppose an MMT guys says “the government should cut taxes by $1B for this year”.

    If we wanted to translate from MMT-speak into monetarist-speak, this is what we would get: “The Fed should permanently increase the money base by $1B. And since this would increase the present value of the Fed’s profits by $1B, which would increase the present value of government revenue by $1B too (since the government owns the Fed), the government should spend that $1B by cutting taxes by $1B for this year. Just helicopter the $1B.”

    Same thing, in different languages.

    My pragmatic distinction between monetary and fiscal policy is: if the central bank does it it’s monetary; and if the government does it it’s fiscal. But yeh. The two are linked if the government gets the central bank’s profits.

    Which means it all comes down to “fiscal dominance” vs “independent central bank”.

    With an independent central bank, the government’s budget has to conform to the profits the central bank chooses to earn.

    With fiscal dominance the central bank’s profits have to conform to the government’s budget.

    I normally assume an independent central bank. Fiscal policy has to conform to monetary policy, sooner or later.

    MMT guys assume fiscal dominance. Monetary policy has to conform to fiscal policy, sooner or later.

  50. Auburn, you write:

    “But the case is made clearer when considering the example of a budget surplus environment. Lets say that the Federal Govt ran a budget surplus of $1 dollar ever year for 30 years.”

    OK, that’s $30 after 30 years. (BTW, I do account for a surplus in Ex #11 in the more complex BSs with the variable “Ut”)

    “Almost every outstanding bond would have matured.”

    OK, but why do you write this?:

    “That exact amount of money would have been “paid back” by the Fed to the bond holders as either reserves or bank deposits.”

    First of all it’s the TSY not the Fed that’s responsible for paying the bond holders back!… including the Fed (the Fed is a bond holder and the Tsy must pay it the principal back… that principal (unlike the interest) is NOT remitted to Tsy…. instead (since it’s going to the Fed) it’s annihilated (debited from the TGA).

    “And yet, Congress would never have had to acquire $16T in deposits from the private sector to redistribute to the bond holders.”

    OK, sure the debt ceiling would not have to be raised… the old debt would be replaced as a liability on Tsy’s balance sheet with the new debt… you could imagine this is done incrementally over the 30 year period.

    So what are you getting at?

  51. There is a gap in the logic of this that I think you have argued both ways at times. Specifically regarding QE I think you have gone both ways without taking a firm position. If QE is a pure asset swap then the only effect it has is through portfolio rebalancing. The connection between the IOR (or any other MZM entity) is completely unrelated to anything the Fed can control. Look at the data for the yield curve. The short end I would argue is highly correlated with Fed policy. Tsy maturities 1 yr and less correlate extremely well with the FedFunds (and IOR) rates. But the 10yr (and mostly the 7 and 5 yr) yields, have a much looser correlation with Fed policy (for rates it can control). Huge increases in very short rates have led to about 4 inverted (or completely flat) yield curves with long (10yr here) yields rising since the early 1980’s, but otherwise long yields have been in a steady, almost linear decline for 3 decades.

    Looking at the data, QE has resulted in higher long yields. During each period of QE, long yields have increased (higher at the end than the start). Short yields have been completely flat.

    So the data clearly show that QE, if it has an effect, raises long yields (the data is a fact). But you have been against QE because of it’s portfolio rebalancing effects. Yet the data show that periods of QE have increasing long yields, which all other things being equal (they are not) pushes the portfolio balance towards debt.

    What really is your position on QE?

  52. Bernard: my memory is failing, but I think Lars Svensson (as a board member) advocated the central bank of Sweden do exactly that. And Silvo Gessel wanted to do the same thing for currency with stamped money in the 1930’s. Everything new is old.

  53. With a budget surplus the govt wouldn’t actually pay back bondholders by printing money. It would pay from its revenues. The govt is not authorized to spend funds it doesn’t have. It is not presently authorized to do so. You are theorizing that it can when it legally cannot.

    Look, this is not an MMT thread and I am going to ask you kindly to stop talking about MMT here. I am sick and tired of writing posts that have nothing to do with MMT and then having MMTers swarm the comments arguing about their theory. I shouldn’t even have mentioned the letters in this post, but still, this post has NOTHING to do with MMT and their theories so please don’t hijack the thread and try to turn it into a MMT discussion. I ask you kindly to please do it somewhere else. Thanks.

  54. I think that’s a good way of looking at it. But the implementation of Fed policy still comes down to the details. That’s where the rubber meets the road and I guess I just don’t have as much faith in the expectations channel as you guys do. So I still want the real meat from a transmission mechanism, but I realize you guys don’t think it’s totally necessary. Still, if David and I were running the show together we’d have implemented NGDP targeting with tax cuts yesterday! :-)

  55. Good piece, Cullen.

    This ongoing back and forth with Krugman reminds of Bernard Lietaer’s claim…

    “Paul Krugman told me personally that it was totally crazy to talk about the money issue…we were both from MIT … what he told me: ‘Didn’t they tell you? Never touch the money system! Never touch the money system! You can touch everything else, never touch the money system.’ … And the reason?…You will not be invited to the right places and you can kiss goodbye the Nobel…”

    http://www.youtube.com/watch?v=AiSM5ozdmgE#t=02m14s

    Well, now the Krugman has his Nobel, maybe he’s willing to touch the money system?

    It’s a bit sad really. But better late than never!

  56. Well, if over a 30 year period, the Govt issued no new bonds due to being in a sustained $1p.a. surplus, almost every bond would have matured in that 30 years. Every bond holder would have gotten their original deposit back (plus interest) and Congress would have never had to tax $16 T in surplus in order to get the bank money with which to redistribute around back to the bond holders.
    This is a serious flaw in your rationale. If $16 trillion in bonds can be redeemed with only $30 in surplus funds over a 30 year period. Then deficit spending is adding new money to the system. Thanks for the back and forth Tom, I enjoyed it.

  57. You mean, by never once mentioning MMT, I’m turning this into a MMT discussion? Thats peculiar. I read your stuff every day, thanks for that. I am sick and tired about reading about Vincent’s ridiculous Hyperinflation hyperventilating, but you seem perfectly capable of carrying on that conversation with him without complaint. All I did was provide a very simple example that your redistribution theory doesn’t explain. Nothing more and I feel that you have really done a disservice to a loyal reading by lashing out at me for no reason.

  58. Krugman eventually calls his antagonists racist. Beware.

    It’s what he does best.

  59. Thats a very helpful way of explaining it Nick and I certainly appreciate there is no one way to understand what is fiscal vs monetary

    I prefer Scott Fulwilers distinction though (and Cullen has said similar things as well) which is that Monetary policy doesnt directly change incomes it changes how far your current income can go via interest rate changes while Fiscal policy directly affects incomes via redistribution or direct purchases of goods or services by the public sector.

    This;

    “With an independent central bank, the government’s budget has to conform to the profits the central bank chooses to earn.
    With fiscal dominance the central bank’s profits have to conform to the government’s budget.
    I normally assume an independent central bank. Fiscal policy has to conform to monetary policy, sooner or later.
    MMT guys assume fiscal dominance. Monetary policy has to conform to fiscal policy, sooner or later.”

    is helpful in clarification of fiscal/monetary but monetary obviously is undemocratic. I want to hear all you MMers state that explicitly. You are for a small group of unelected people determining the monetary fate of billions. See how popular that message is. I think you guys want us all to think that this monetary policy you push for IS the best choice for modern democracies yet what you advocate is for modern democracies to become explicitly less democratic.
    Just let Chuck Norris decide what to buy and for how much with no discussion within our admittedly imperfect system.

  60. I have not had time to digest this excellent post, but it’s another example of not only a great post from Cullen, but the comments are mostly excellent too, in terms of helping us all grasp the concepts.

    I would like to hear some opinions on the last sentence in the article:

    * We can quibble over QE’s impact on long rates, but I’ll take the SF Fed’s word for it that it’s “negligible”.

    The Fed has not started “tapering” yet and has essentially reiterated what they’re looking for in the economy before they begin actual tapering.

    Seems to me that the fear of tapering has hit the bond market hard. Over-reacting? Perhaps this is what the banks want, since the Fed is essentially the banking industry’s bitch.

  61. Auburn,

    It’s not me “lashing out”. I asked you kindly not to turn this into a MMT thread. That’s all. I’ve had all these debates with MMT people before and they all end the same way. Just trying to nip this one before it goes that direction.

    Thanks for understanding.

    Cullen

  62. You don’t need to track anything, just change the exchange rate between paper money and electronic money at the desired rate (this is Miles Kimball’s plan for getting rid of the 0% floor, but it’s equally a plan for paying positive interest on currency).

  63. Okay, but there’s still a problem with Nick’s idea. Increasing the MB doesn’t increase the Fed’s profits by $1B. I think he might have had a bit of a brain fart there….

  64. Cullen: Yeh, I should have been more precise there. Increasing the MB permanently by $1B increases the present value of the Fed’s profits by the same $1B, if the MB pays 0% interest on the MB and we ignore the costs of printing new notes to replace worn out notes etc. Because the Fed buys a bond with PV of $1B, and doesn’t have any extra costs.

  65. Let to me try to simplify. If the price level jumps by x% (i.e. unexpected inflation), then the central bank has a profit of x%*M0. This has nothing to do with seignorage…the profit comes from defaulting on the CB’s price stability promise. An interest-paying, zero-seignorage bank enjoys the exact same profit.

  66. Greg: “…but monetary obviously is undemocratic. I want to hear all you MMers state that explicitly.”

    OK. Here you are:

    Canadian (and US?) monetary policy is “undemocratic” in roughly the same sense that the Canadian (and US?) supreme courts are “undemocratic”. Both are run by powerful people who are not directly elected but are appointed for long terms by people who are elected.

    And that is one reason why we want monetary policymakers to follow a rule (like inflation targeting or NGDP targeting) rather than actively using their discretion to do whatever they feel like doing at the time.

  67. Ah nevermind….that’s a real brain fart. The profit isn’t x%*M0 but rather x%*reserves (gold, foreign currency).

  68. When the Fed buys bonds they’re issuing a liability and taking on an asset. The Fed’s balance sheet doesn’t change in terms of total net worth unless they’re earning more on the bonds they own than the cost of the MB. This is why I keep saying that QE isn’t real “money printing” in any meaningful sense. The Fed’s just performing asset swaps with the pvt sector which changes portfolio composition and has all sorts of intangible side effects, but it’s not the same as having the US Tsy print up a T-bond and issue it or having the Fed buy bags of dirt in exchange for new pvt deposits (which would theoretically be liabilities of the Fed). Said differently, monetary policy is most effective when it takes the form of real fiscal policy!

    Are we just talking past one another or do you see where I am coming from there?

  69. Cullen: I think I see where you are coming from.

    Monetary policy has two effects:

    1. The income/wealth/fiscal/Pigou effect. If you print up a load of $100 bills, and helicopter them out, or give them to the government which hands them out, everyone who gets one is $100 richer, so they want to spend more.

    2. The substitution/portfolio composition effect. We don’t want to hold more than a fraction of our wealth in $100 bills (and that fraction depends on stuff like interest rates and expected inflation and income). If the actual fraction gets bigger than the desired fraction, we start spending more.

    The first effect is small, unless you are talking really massive percentage increases in the money supply. Because base money is only a small percentage of annual GDP, and an even smaller percentage of total wealth (including human capital).

    The second effect is where nearly all the action is, at least in normal times. It explains how the Bank of Canada, which is much smaller than the big commercial banks combined, and whose balance sheet is tiny compared to total wealth, can control the whole economy to keep inflation on target.

    And the current debate is over whether the second effect can still be made to work in a liquidity trap.

  70. Nick, is the second effect the HPE? Sumner, just days ago, stated that at the ZLB the HPE is “very weak.”

    “PS. I can see how someone might have assumed I relied on the Pigou effect in the past, as I’ve argued that QE probably has a small effect above and beyond the expectations channel. Cash and bonds are not perfect substitutes, even at zero rates. But I’ve never emphasized this channel, and in any case it’s based on a (very weak) hot potato effect, not the Pigou effect.”

    http://www.themoneyillusion.com/?p=22948#comment-267116

    Actually… I guess he kind of says the HPE is “based on ” the Pigou effect there, right?

    Whatever the case, I was a bit astonished at this and asked Scott if there was more than one HPE. He said “No.” and went on to state:

    “In a sensible system the base money is endogenous. You set the NGDP target, and the public tells you how much base money they want to hold. I’m all for that.”

  71. Nick, they have been following rules for years via inflation targeting. You now think the rule should change but you really have ZERO clue what rules they should follow to get the effect you desire (increased NGDP). Should it just QE infinite? Should they buy bags of dirt? Should they shoot dollar bills out of bazookas into shopping malls. It honestly sounds like you guys are just flailing yelling at Bernanke “just do something extreme!!”

    You have said that everything should be considered (maybe it was Sumner who said that) but thats no rule at all! Do everything/anything to get our price level to rise?? Thats not policy. Thats not rules. Thats anything goes. Its actually 180 degrees from rules.

    Funny you choose the Supreme court their popularity is at an all time low.

    Not that popularity is THE metric to follow in and of itself but if you want to create bad expectations from the people who are actually necessary for our economy to function, the 10s of millions of consumers, tell them your ready to appoint Ben Bernanke central buyer if need be.

  72. Tom, that quote sounds very similar to this one which summarizes Scott’s entire philosophy:

    “The Fed does monetary policy by setting expectations, and then providing the amount of base money for the monetary base market to be in equilibrium when NGDP expectations are on target. It’s that simple.”

    I have no idea why he feels the monetary base is such a big deal when it clearly only performs a facilitating function. However, if they drop cash from helicopters, then that is a big deal. Nick Rowe apparently disagrees (see his number 1 example above) but increasing incomes/wealth is the whole idea!

    It is his number 2 example that is of lesser importance. It appears Nick has it backwards.

  73. Geoff, you write:

    “I have no idea why he feels the monetary base is such a big deal when it clearly only performs a facilitating function.”

    I don’t either when we pile it up in reserve accounts (QE) at banks, but Scott seemed to be saying (in my quote from him) that this piling up isn’t necessary… let’s just let the market determine how much is needed (i.e. “endogenous”). I specifically asked him (in that comment) “So you’re saying lets just go back to ER = 0 and let the Fed provide base money as needed: no more no less just like they did prior to 2008… only this time they make NGDP explicit?” and he replied “Yes, let’s do that” and then the rest of the quote you see there.

    BTW, I know I’ve pestered you enough w/ this stuff, but did you see how your spreadsheet has evolved? Thanks again! (If you hate it and demand I take your name off, just let me know! ;)

  74. I did see it! Did you receive my reply? It was delayed as I’ve been offline for a while.

  75. Here’s what I wrote actually:

    “OK, so if you agree with David here, and HPE is “very weak” when we’re at “zero rates” then why not ditch the QE and the ER > 0, and just promise that base money will be made available when needed. And then… if we do that, how is that different than what we have been doing prior to 2008? Except that now there’s an explicit NGDPLT?”

    http://www.themoneyillusion.com/?p=22948#comment-267056

    That was my “last point” and Sumner’s sentence just prior to my quote of his above (several comments up) was:

    “And I certainly agree with your last point.”

    Elsewhere he even admits QE has been a “mess.” So all we need is a promise from the Fed they’ll keep doing what THEY’VE ALWAYS been doing, re: supplying “base money” as need … plus an NGDP target. :O

  76. Sorry, you’re wrong about banks not lending out reserves.

    A bank loan can be completed in one or more of the following ways:

    1) The lending bank creates a deposit for the borrower.
    2) The lending bank creates a deposit for the borrower’s designated recipient.
    3) The lending bank delivers federal reserve account funds to another bank, which creates a deposit for the borrower.
    4) The lending bank delivers federal reserve account funds to the bank of the borrower’s designated recipient, which creates a deposit for that recipient.
    5) The lending bank delivers federal reserve account funds to the borrower bank.
    6) The lending bank delivers banknotes to the borrower.
    7) The lending bank delivers banknotes to the borrower’s designated recipient.
    8) The lending bank delivers banknotes to the borrower bank.

    You are focusing only on method 1, which is common, but so are methods 3, 4 and 5. For example loans made for specific purchases and payments from credit cards and other credit lines are typically completed by method 3.

    While it’s true that healthy banks in healthy economies don’t typically check their own supply of reserves before approving loans, that’s only because such banks are confident that reserves are readily available. An unhealthy bank that has no access to interbank lending and no understanding with the central bank that reserves will be made available to cover its lending is forced to stop lending.

    Even healthy banks in healthy economies are very dependent on the general supply of reserves. They include the price of borrowing reserves in their lending rates. Typically the central bank has promised that enough reserves will be made available to keep the cost of borrowing reserves near some target rate, albeit often subject to frequent revision. Banks after all do take into account reserve supply, broadly understood, in all their lending decisions.

    As for the S&P quote, I suspect you misrepresented it. What he probably meant is that banks don’t and can’t lend reserves outside the banking system. Actually they typically don’t, but they can and sometimes do, in a loan completed by method 6 or 7. Such loans are typical of less advanced economies where large transactions are made with banknotes. But they can and do happen even in the US from time to time.

  77. Cullen

    Well I be Damned never thought I would see the day Krugman would offer a Olive Branch.

    Great Job, you actually got Krugman to say you were right, he even put a ! point after it.

  78. Yeah, but you just skipped the interbank settlement step. Your #’s 3, 4 & 5 are the same as “bank A creates deposit at Bank A account for borrower B who then pays for burrito at restaurant with account at Bank B thereby requiring Bank A to settle in reserves with Bank B”.

    All you did was skip a step there to give the appearance that the reserves create the deposit. Yes, banks settle payments with reserves and lend them to ONE ANOTHER in the interbank market.

  79. Confused (no surprise, huh) but in those examples the bank is lending out central bank reserves. (?)
    Are those the same as bank reserves held at the Fed?
    So bank A is not lending out its own reserves, but somebody else’s reserves? But even if there are no reserves available, the central bank will create some?

  80. Tom, I wrote this out earlier today:

    http://pragcap.com/failing-the-great-recession-test/comment-page-1#comment-152260

    It may help see the argument.

    Basically “lending out reserves (to anything other than another bank)” amounts to cash advances… and that’s if you ignore the technicallity that reserves by definition cannot leave the banking system (so where can the go “out” to?).

    More at the end of this:

    http://brown-blog-5.blogspot.com/2013/08/banking-example-11-all-possible-balance.html

    Plus a cash advance is really a two step process, blurred into one:

    1. non-bank takes out loan and receives deposit
    2. non-bank exchanges deposit for cash

    That’s the same as increasing both independent variables L and C on my embedded spreadsheet in the above link.

  81. Johnny, you write:

    “…in those examples the bank is lending out central bank reserves. (?) Are those the same as bank reserves held at the Fed?”

    Yes

    “So bank A is not lending out its own reserves, but somebody else’s reserves?”

    Bank A? I didn’t see “bank A” in there … well in any case a bank can lend out it’s own reserves or it can borrow them and transfer them if it needs to. I think that’s what Tom is getting at.

    “But even if there are no reserves available, the central bank will create some?”

    Yes, absolutely.

    Everything Tom writes there can be summed up with a couple of is closing remarks:

    ” What he probably meant is that banks don’t and can’t lend reserves outside the banking system. Actually they typically don’t, but they can and sometimes do, in a loan completed by method 6 or 7.”

    YES! Absolutely we’re talking about loaning “outside the banking system.” Nobody disputes that banks can loan reserves inside the system. As for his method’s 6 & 7… to get super nit-picky, reserves, by definition can’t leave the banking system. Base money can, so I’ll grant him that (in the form of cash), but now we’re talking cash advance as he refers to. But that’s not terribly different that someone (a non-bank) just exchanging a bank deposit for cash: that’s not controversial… no one disputes that can happen.

  82. Thanks, I get all that.
    Probably some people don’t understand that the central bank will provide reserves if none are available.
    If you don’t know that, then you see the reserve money as a pool of lending money and figure the banks would rather lend it than let it sit.
    I guess, though, I don’t get the caveat that the money can’t leave the banking system.
    If I borrow money (deposit created) and buy something by transferring that deposit to your bank, the money doesn’t leave the reserve system, but didn’t we just conduct business that we couldn’t have done without the initial loan.

  83. Johnny, you write:

    “I guess, though, I don’t get the caveat that the money can’t leave the banking system.”

    I think I said the reserves can’t leave correct? That’s just a definition:

    Fed deposits held by banks are reserves. And cash at banks is reserves.

    Fed deposits not held by banks are not reserves by definition (e.g. the TGA = Tsy’s Fed deposit). Cash not held by banks is also not reserves either.

    Since outside money = (Fed deposits) + (cash in circulation (i.e. not including cash held at the Fed))

    Then we could simply state that outside money held by banks is by definition reserves. Outside money NOT held by banks is NOT reserves.

    Make sense? I’m not saying that either Fed deposits or cash can’t leave the banks: that outside money certainly can! It’s just that when it does so, it’s not longer defined as reserves: it’s still outside money but not reserves. And of course Fed deposits can ONLY go to Fed deposit holders, of which there are only a few (especially rare are non-bank Fed deposit holders).

  84. Thanks for posting… I used to go to Keen’s quite a bit… and in fact almost went over again to see if caught wind of any of this. You made it a lot easier.

  85. Geez, either you just said that reserves can’t leave the system because once they do they’re not reserves, or you’re just yanking my chain.

  86. That’s what I said.. I’m not yanking your chain… but with WAY too many words. Ha! … sorry. went overboard with that I guess. ;)

  87. Roche & Krugman, debating.
    The fiddlers are playing whilst Rome burns.
    So much pain ahead for these two.

  88. “Probably some people don’t understand that the central bank will provide reserves if none are available.”

    Sorry don’t mean to interject too much considering this is not a conversations I’ve participated in.

    If the Fed explicitly sets a policy interest rate target on Fed Reserves (Fed Funds Rate), then the Fed must be ready to provide fed reserves at that policy rate. If the Fed does not, the bank that is short of reserves will need to offer a higher rate of interest on reserves in order to entice the banks with surplus reserves to make a loan to them.

    Since the Fed is the monopoly supplier of Fed reserves, they can either control the quantity of reserves banks hold or the price (interest rate) on reserves, but not both. As the Fed targets the interest rate, the quantity of Fed reserves held by the banks fluctuates and is determined by the needs of the closed loop banking system.

  89. http://cowles.econ.yale.edu/P/cp/p02a/p0205.pdf

    James Tobin argues that banks do create money, simply by making loans and crediting the borrowers bank account )see pdf above).

    In his latest piece for the New York Review of Books, Paul Volcker says that interest on excess reserves (the Fed pays 0.25 on IOER ) has “sterilized” the large increase in bank reserves.

    “The banks can’t lend out the reserves” sounds weak in the face the above arguments.

    That said, I do like the idea of the Fed buying bags of dirt for $100, or a national lottery (we can just piggy back on state lottery mechanisms) that pays out more than it brings in, and pays out in cash printed by the Fed (okay the the Bureau of Engraving actually prints the stuff, but buy know what I mean).

    I propose small winners, of $100 or so, say (on average) for every $75 wagered. A max on annual winnings at $10k, as winners have to provide their SS numbers. No more than $50 in tickets can be bought at any location in any day. This prevents large-scale packaging or bets.

    Well, it is a slightly better idea than bags of dirt….

  90. I didn’t skip any steps. I laid out all the possible ways a loan can be completed. I didn’t do anything to “give the appearance that the reserves create the deposit.”

    It’s apparent now that you actually agree with everything I wrote. In other words, you have misstated your own case. You said that banks don’t lend out reserves, but now, when I have laid out very plainly that banks do in fact lend out reserves, you admit that they do. Now you’re saying instead that banks don’t lend out reserves outside the banking system. That’s a very different point.

    This time you’re close to the truth, but still not technically correct.

    Base money is central bank reserve account funds and banknotes. They are interchangeable on demand.

    Reserves is central bank reserve account funds and banknotes held by banks. Central governments and in the US some GSEs also hold central bank reserve accounts, which may or not be counted among reserves depending on who’s doing the counting.

    When a bank makes a loan and completes it by delivering banknotes to the borrower, it has lent out reserves outside the banking system.

    Reserves leave the banking system all the time, every time anyone withdraws banknotes. Likewise banknotes deposited at banks increase the volume of bank reserves. There’s a constant, enormous flow in both directions, and a gradual net outward flow of reserves to the banknote-holding population.

    I think what you are trying to get across is that lending doesn’t *necessarily* result in a decrease in reserves, and very rarely results in a one-for-one decrease in reserves. But I think almost everyone already knows that.

    The standard textbook explanation of how lending consumes reserves works through two routes: increased lending is presumed to result in increased activity and increased demand for banknotes, and increased lending is presumed to increase required reserves. Note that the latter route does not decrease total banking system reserves, it only decreases excess reserves.

  91. Okay, then, as I wrote originally, you’re wrong. Banks lend out money to any kind of party. They often deliver the money in the form of reserves, usually by delivering central bank reserve funds to another bank, and rarely by delivering banknotes to a person or non-bank. Only when the borrower is a bank can the lender bank be said to be lending its reserves to another bank. Whenever banks make loans to non-banks that are completed by immediate delivery of reserves, either central bank reserve funds or banknotes, the bank has lent out reserves to a non-bank.

    I’m aware of much more of the actual accounting than you seem to be, and I’ve laid it out for you very clearly. You’ve pointed me to an example of my case no. 1. You haven’t denied that all my other cases are also equally realistic. You’ve only made a weird statement that I “skipped a step” with cases 3 through 5, which is hard to understand and anyway wrong. My steps 3 through 5 describe actual bank practice, with no step skipped.

    By the way, the case you pointed me to is unrealistic. A bank with zero deposits will not lend. Such a bank would not have access to the interbank market, nor would it have access to the central bank window. The interbank market is treated by healthy banks as a practically sure supply of reserves, but that’s because the bank has built up a sound reputation and certainly has a considerable deposit base. Also, central bank windows are not typically relied on to cover ordinary lending. Central bank windows are fall-backs used in emergencies, when interbank supply is cut or highly priced, to cover deposit withdrawals and credit line payments, not usually new loans – but that’s a rule of thumb and there are exceptions.

    What we’re really getting down to here is that you’re trying to lecture people on how the banking system works, but you don’t seem to know very much about how the banking system works. You need to step down from such absolute statements as “banks don’t lend out reserves.” You don’t seem to understand what the word “reserves” means. You seem to think reserves refers only to central bank reserve account funds. That’s not so. Reserves also includes banknotes held by banks.

    There is a valid point that you’re trying to make, but you’re misstating it.

    The point is that bank lending, when looked at from the perspective of the total banking system, creates deposits out of thin air, and does not much reduce total system excess reserves. Total system reserves can only be reduced by net withdrawals of banknotes from banks, which are only loosely related to the rate of net new lending and on average always far less than one-to-one. Also, excess reserves are reduced only very slightly by net new lending, because reserve ratios are small and sometimes zero.

    That is the actual, accurate situation I believe you’re trying to describe.

    But your attempt to illustrate this point by narrowing in on what happens when an individual bank lends is a failure. You’re using an idealized example of a bank loan to make absolutist statements about how bank loans work that don’t hold true in many situations.

    You need to abandon that tactic and widen out your focus to the total banking system. Above all you need to educate yourself before you try to educate others.

  92. I’ve never said banks don’t lend reserves to one another. That’s just an obvious fact of banking for anyone who understands the basics. You’ve taken a very basic and obvious fact and blown it up into some sort of grand argument that was never being contested.

    So I don’t really see what you’re trying to achieve here. There’s absolutely nothing you’ve stated here that hasn’t been stated 100 times before you came here. I know you’re new, but I never said banks don’t lend reserves to one another and I never said reserves didn’t include cash. Those are very very basic facts that shouldn’t need to be hashed out for anyone who understands the scope of the discussion here. All of the things you discuss have been touched on either here or on Tom Brown’s website where the accounting is laid out in excruciating detail. If you take the time to look around you’d probably find that you don’t disagree with anything I’ve said….

  93. Tom, do you disagree with this definition of reserves:

    “Bank reserves are banks’ holdings of deposits in accounts with their central bank (for instance the European Central Bank or the Federal Reserve, in the latter case including federal funds), plus currency that is physically held in the bank’s vault (vault cash).”

    In other words reserves are defined as base money AT BANKS! Base money not at banks is not reserves by definition. Reserves can’t be “loaned out” to entities outside the banking system by definition.

    Banks CAN lend out base money to entities outside the banking system via cash advances. That’s it. They can ONLY lend reserves to each other.

    And even the cash advances argument is a little iffy: what actually happens? Is a deposit created (via the loan) and then swapped $ for $ for cash? I don’t know the technical details, but it hardly matters. Reserves NEVER leave the banking system by definition. Base money can via swapping bank deposits for cash.

  94. Banks most certainly CAN lend out reserves: to each other. By definition reserves are base money **at banks**: reserves can in NO circumstances leave the banking system, because base money outside banks is NOT reserves by definition.

    The ONLY way that base money leaves the banks into the non-bank private sector is via exchanging bank deposits for cash, or via a cash advance (if a cash advance is actually different than this: I don’t know if a cash advance is really a two step process or not: 1. loan creates/credits a deposit 2. deposit is immediately swapped for cash.

    But either case base money leaves the banking system into the non-bank private sector by one basic mechanism: bank deposits are swapped for cash. If deposit holders don’t want to make this swap, then it doesn’t happen.

  95. The Fed could induce conversion of reserves into currency by charging banks to hold reserves (i.e. negative interest). The immediate effect would be to cause a currency shortage, since the Fed wouldn’t be able to print $100 bills fast enough to satisfy the demand. Important to note, however, that this currency would be just as inert as reserves, for as long as 0% is considered an acceptable interest rate.

  96. Base money can also leave the banking system by either going back to the Fed where it’s destroyed (either Fed deposits or cash) or to a non-bank Fed deposit holder, such as Tsy.

  97. That would be a return to normal, because before 2008 banks were taxed on their reserves; interest was 0% and rates were positive.

    In countries without any reserve requirement, a negative interest rate on reserves is not a bank tax, since holding reserves is voluntary.

  98. The $2 trillion is almost all excess reserves. These are the reserves that would leave the banking system if IOR was negative. Only the required reserves are a “tax” – which could be eliminated if IOR was set differently for required and excess reserves.

    I’m not suggesting this is a good idea. Without a plan to get rid of the 0% floor, it doesn’t make much sense. But it’s technically possible to force banks to “lend out” reserves.

  99. So you’re advocating a 2 trillion dollar bank run as your solution to fix the economy?

  100. Making IOR negative enough might cause depositors to chose to exchange their deposits for cash since they’d have to pay for the privilege of keeping a deposit (and thus a new business might spring up: cash warehouses that offer credit… but this isn’t necessarily a bank!).

  101. Okay, I’ll let it go. But I think what you’re doing is weaseling out of what you claimed after I proved that what you said isn’t true.

    You said that banks don’t lend out reserves. I proved to you that they do.

    You changed your position to banks don’t lend out reserves except to each other. I showed that they do. When a bank makes a loan to a non-bank and delivers reserves to a bank on behalf of the non-bank, the bank has lent reserves to a non-bank.

    You replied that we agree. So apparently, your readers are supposed to be satisfied that you know how things really work, even though you’re struggling to fit words together into a factual sentence?

    The ironic thing is we do seem to agree in the endogenous versus exogenous money debate. Krugman is completely wrong and just plain silly to point to Tobin’s argument that the population only holds as much deposits as it wants to. That’s a terrible model of how banking works. But you won’t be able to overcome that bad model and show that Krugman’s wrong as long as you’re factually wrong statements like “banks don’t lend out reserves” that miss the point.

  102. This is a cute but wrong technical argument. Yes, reserves cease to be reserves as soon as they are transferred from the possession of a bank to a non-bank. Reserves don’t exist outside the banking system by the usual definition. That doesn’t change the fact that the bank has lent out its reserves. The bank had x reserves, the bank lent y, the bank now has x-y reserves. Thus the bank lent out y reserves.

    Reserves do leave the banking system whenever payments are made to the central bank or any non-bank that banks at the central bank (in the US: Treasury, Fannie and Freddie). At the point they leave the banking system they cease to be reserves. Reserves that leave the banking system and cease to be reserves have of course left the banking system.

    You won’t make yourself better understood by saying things that are literally wrong and then when it’s pointed out countering that you have your own special definitions of certain words.

  103. No, what I did was write a BLOG POST about banking and I used oversimplified terms and examples to show a general point. Then you pounced on nuances in operational realities (nuances that have been discussed here in vivid detail) and proclaimed yourself the master of the banking universe.

    Look, I write a money for dummys type blog. And at times that results in very simplified posts. If I covered every detail of banking in a blog post you wouldn’t even come here to read it. It would be pointless. I realize this and at times I have to oversimplify things to get a broader point across. Of course banks lend reserves to one another (I’ve written about this FOR YEARS with regards to the overnight rate and impact of reserves), but that has nothing to do with the point of this post and you know it. Everyone who understands reserve accounting knows that. But I don’t have the time nor the space to cover the nuances of reserve accounting in a brief banking post.

    And now you’ve come here to beat your chest and proclaim that I don’t know what I am talking about because you read one blog post. Based on your traffic here you just started reading this website and you don’t know the first thing about me. But you’re quick to jump to a bunch of conclusions upon reading ONE brief blog post.

    If you want to contribute to the knowledge of the community here then feel free to add to the blog post with your own comments instead of coming here and attacking a short post with the declaration that “you’re wrong about…” There are much more constructive ways to go about getting your message across. Like, “Hey Cullen, first time reader. I don’t think your post is detailed enough to communicate the message you’re trying to convey. You might want to add that banks can do XYZ….”

    Instead, you came here with a defensive attitude trying to attack someone who has covered every detail you proclaim to be the expert on….And in doing so you’ve accomplished absolutely nothing, but wasting a bunch of time. Why bother? Why strut around the internet with such an attitude? It’s pointless and now look at all the time we’ve wasted because of it….

  104. Tom, I for one never imagined that when Cullen, or Krugman, or Sumner or Schiff for that matter, speak of “lending out reserves” that they’re talking about inter-bank lending. I’ve had numerous discussions about this with Scott Sumner on his blog, and neither one of us made that assumption. I by no means this is unique to Cullen here.

  105. Sorry, Cullen, but you’re still weaseling out of your mistake. You’re still not admitting that banks do lend out reserves, not only to each other, but to non-banks. I didn’t pounce on a nuance, I pounced on a core misconception in your belief system and pedagogy.

    I don’t think your readers are dummies nor do I think blogs are an excuse to be misleading. Matt Klein made the same mistake as you and more the other day and I ripped him for it as well. Go read his article if you want an example of how far people go wrong when they read what you call “oversimplified” statements that banks don’t lend out reserves. He’s going to the point of telling people that aggregate lending doesn’t depend on the supply of aggregate reserves. (So how did the Fed control the interest rate until 2008, by magic spells?)

    There are a lot of people out there taking you literally when you say that banks don’t lend reserves. It’s misleading. Banks do lend reserves, and not just to each other. Telling people that banks don’t lend reserves isn’t the right way to explain why aggregate banking system reserves are only slightly reduced relative to aggregate net bank lending. Tell people the truth: lending consumes on average only a small amount of excess reserves, through increased banknote withdrawals and increases in required reserves.

    That process of consuming excess reserves by spurring banknote withdrawals and increases in required reserves is what economists are referring to when they point to excess reserves and say, look, there hasn’t been enough lending to consume those excess reserves. Of course it would take a ridiculous amount of lending to consume $2 trillion of excess reserves, but there’s nothing false about saying that there has been too little lending to consume it.

    But because ideas like “banks don’t lend out reserves” have such currency, we have a columnist on Bloomberg telling people that the economists are wrong and don’t understand how banking works. That’s not true. The economists did understand, and it was Klein who was mistaken.

    Again, this isn’t a tiny technical debate. This is about the core of your belief system and pedagogy. Contrary to what you teach, the supply of reserves is absolutely crucial to lending. We just happen to have so much more excess reserves in our banking system than could possibly be consumed by lending anytime soon that marginal changes in the reserve supply don’t matter to banks’ ability to lend. That wasn’t true until 2008, and at some point in the future, the situation could revert back to what used to be considered normal, when aggregate excess reserve was typically very tight and Fed manipulation of the reserve supply had obvious, immediate impacts on interest rates and lending.

  106. “when aggregate excess reserve was typically very tight and Fed manipulation of the reserve supply had obvious, immediate impacts on interest rates and lending.”

    Which is why the Fed adjusts the banking reserve levels w/ repos/reverse-repos to precisely hit and maintain their FFR target at all times (w/o wild fluctuations) in reaction to what the private sector’s needs are.

  107. No, we’re not talking about interbank lending. We’re talking loans to non-banks. I don’t know about Sumner or Schiff, but what Krugman means is the same thing I mean: when a bank lends, it delivers up reserves to somebody. The fact that the reserves usually stay within the banking system is a different point. The fact that the reserves are usually delivered to another bank for credit to the account of the non-bank borrower or his designated recipient does not make it an interbank loan.

    I don’t think it’s so hard to be accurate about this. All you need to say is:

    When banks lend, the supply of deposits increases by a like amount. Nearly all of the money they lend stays within the banking system, except for on average a small portion that gets withdrawn as banknotes. The increase in deposits also increases the amount of required reserves banks must hold, but only by a tiny fraction of the increase in lending. In some banking systems, like the US until 2008, the central bank keeps the supply of excess reserves quite tight, which allows it to spur or suppress lending by slightly increasing or decreasing the supply of reserves. In other banking systems, like the US since 2008, the central bank ensures that the supply of excess reserves is far greater than banks could consume, and the central bank instead controls interest rates by changing the interest rate it pays on reserves.

    Is that so difficult to understand? I really don’t see what’s to be gained by trying to “simplify” that story into one in which banks “don’t lend out reserves.”

  108. You do realize there are countries without reserve requirements right? And that these banking systems make loans just fine, right? I am now beginning to realize you actually don’t understand my point about reserves not being lent out at all. No one is “weaseling out” of anything. You actually don’t understand the point. Like Krugman, you think that this is some sort of temporary environment and that once we don’t have the excess reserves, that banks will then be constrained by their reserves. That’s just wrong. Sorry.

    You might want to read the following pieces:

    http://monetaryrealism.com/loans-create-deposits-in-context/

    http://monetaryrealism.com/the-accounting-quest-of-steve-keen/

  109. Okay, now it’s clear that you actually don’t understand how this works despite your lecturing.

    When the banking system makes loans and needs reserves to meet requirements the central bank must supply them. There is no central bank choice in the matter. Banks make loans and find reserves after the fact. If the central bank hasn’t provided them to meet requirements then it provides them. it doesn’t tell the bank to cancel the loan it made. That’s not how things work. You think it’s the other way around. You think the central bank can starve the banking system of the reserves it needs to meet requirements. That’s now how the order of operations works. The central bank doesn’t starve a solvent bank from meeting its legal requirements. And bank loan departments don’t check aggregate available reserve before making loans. Again, this is not how things work at the operational level.

    Also, the Fed did not control lending via interest rates pre-2008. If it could have choked off lending just by changing interest rates (by changing reserves) then this chart wouldn’t look like this:

    This chart shows, without a shadow of a doubt that the Fed couldn’t stop the borrowing binge before the housing boom/bust by changing rates. There is NO negative correlation between higher Fed Funds Rate and the level of borrowing. NONE. So your whole framework is based on the same flawed textbook model that Krugman is using. You seem to have a slightly better understanding of things, but it’s still the same basic flawed model that states that we’re in some sort of temporary liquidity trap….It’s not correct.

  110. Cullen: Okay, so you’re going to stick with being wrong and misleading. I’ve explained too many ways how banks lend out reserves to go over it again.

    Your statement about limited excess reserves supply not constraining lending is very wrong. In the the US until 2008, when supply of excess reserves was tight, lending was absolutely constrained by the limited supply of excess reserves. The constraint worked through the interest rate. When the Fed wanted to constrain lending, it raised the target interbank rate and reduced excess reserves supply until the rate rose to the target. Basic, classic monetary policy. There’s no debate to be had: limited system excess reserves constrain lending.

    Tom B: Central banks don’t simply accommodate market “needs”, they regulate credit supply by regulating either the supply of excess reserves or the rate of interest on reserves.

  111. See the chart I posted in my other respnse to you. It 100% destroys the argument you’re making. The Fed can influence a small part of the spread in the bank lending model, but it doesnt equal absolute control over the amount of inside money that’s created. You’re dead wrong here and using the same model Krugman is relying on. Again, see the chart showing how the FFR and loans rose in lock step before the crisis. According to you that shouldn’t have been able to happen, but it did….

  112. “when a bank lends, it delivers up reserves to somebody”

    Yes a net amount of reserves from bank A to bank B in excess of the amount that B delivers to A for exactly the same reasons… generally this net movement is done in bulk only a limited number of times (perhaps once?) over a 24 hour period.. that’s my understanding… and again, that ONLY takes place when the loaned funds are used to spend on a seller who banks at another bank or when the depositor transfers the loaned funds to another bank. When borrower/buyer and seller are both at the same bank, this of course is not necessary… and with just five major banks in the US, this in unnecessary to a significant degree. But what’s left unsaid is the inter-bank borrowing needed to balance this net movement of funds from A to B… from B right back to A through interbank lending (in my two bank world here)…. to cover any overdrafts or to cover reserve requirements deficits on A’s part. In excess of that the Fed is sure to provide these funds through repos/reverse-repos to hit it’s FFR target.

    That’s where the inter-bank lending comes into play here.

    I don’t get what your issue is. Sure, you describe something that I have no fundamental disagreement with, but when considering “lend out reserves” at a macro perspective, the fact that there are more than one commercial banks in existence becomes somewhat of an irrelevant detail that is often glossed over by many many people who’ve addressed this issue. Sumner’s big complaint about that statement, for example, has absolutely nothing to do with your line of argument: He says that banks (by which it’s understood he means “the banking sector”) CAN lend out reserves essentially because they do cash advances (i.e. deposits, created through loans, can be swapped for paper currency). He never mentions this inter-bank stuff you bring up because it’s not really that important from a macro perspective. Hell, Sumner doesn’t even think the banking sector itself is important from a macro perspective!

    I’m skeptical of the claim that a significant number of Cullen’s readership is confused by his statements on this matter. I think a more fundamental misunderstanding might be (amongst the general pubilc) that the Fed has somehow given the banks a lot of money for free that they call “reserves” which the banks are now free (freer that before) to lend out, and that they will as soon as IOR is dropped to 0 (or whatever other trigger happens) and suddenly this flood of reserves will spill out and cause hyperinflation. That, for example, is the view that Peter Schiff (and Vincent Cate) would like the public to believe (so they’ll get scared and invest in his Euro-Pacific Fund… which Schiff runs and thus it must be protected against such a calamity, right?).

    IMO, as far as macro is concerned, multiple banks and the movement of reserves between them to clear payments, transfer deposits, or to borrow and lend for the purpose of meeting RRs, paying back Fed overdrafts, or providing cash to depositors, just isn’t that interesting.

  113. Cullen, I’m beginning to think that one either gets exogenous money or they don’t. And it has nothing to do with intelligence. Dr. K is obviously very smart and so is Tom (although his manners could use some work).

    I’m usually not pessimistic but it appears that this debate is going nowhere.

  114. Tom, perhaps you’d like to check out Nick Rowe on this: Nick is more “neo-classical” than not (he’s certainly not PKE) and he’s come to Krugman’s defense in the past (on banking), and YET here’s what he had to write a few days ago when I misinterpreted some of his statements recently:

    First my statement:

    “Basically one of two [articles Nick wrote] talking about money being endogenous in the short term (between six week meetings of the BoC or Fed) but not in the longer term (like two years out) where inflation targeting makes it exogenous”

    Nick Rowe:

    “Tom: thanks, but that’s not quite right. Under inflation targeting the quantity of money is endogenous in both the short run and the long run. It’s the nominal rate of interest that is exogenous in the very short run (6 weeks or less, for the Bank of Canada anyway), but endogenous in the long run.”

    http://brown-blog-5.blogspot.com/p/links-to-remember.html

  115. Cullen, you honestly believe that when banks make loans “the central bank must supply them. There is no central bank choice in the matter”?

    That’s ridiculous. The central bank has every choice in the matter.

    In a low-excess-reserves system, if the central bank thinks banks are lending to much, it slightly tightens the supply of reserves. It typically merely slows the pace at which it is increasing base money supply. As it does so banks that need reserves to cover lending bid up the rate of interbank lending, and that gets factored into rates charged to non-banks, and they borrow less.

    That said, as a policy choice, the central bank typically precommits to supply as much excess reserves as are needed to maintain a certain base interest rate. But that rate can be changed at the central bank’s discretion.

    If you don’t think a central bank can restrain lending by limiting the supply of excess reserves, what do you think Volcker did, cast voodoo spells?

  116. Well, the data I showed you on the amount of lending vs the FFR during the housing bubble era clearly disproves the idea that the Fed controls the amount of inside money issued via the FFR. This is factual evidence.

    The 70’s recession was not caused by Volcker. It was caused by a 150% rise in oil prices that sucked the life out of consumer. Just like surging oil prices sucked the life out of the consumer in 2008. The main difference between those two environments is that rising interest rates couldn’t stop the consumer borrowing binge despite your claims that the Fed controls lending via interest rates….The 2003-08 period clearly shows that that relationship doesn’t hold….

  117. The chart doesn’t prove what you think it does, by any means. There was a credit boom underway, and the Fed’s gradual tightening was not enough to stop it growing, until it suddenly did. If for example the Fed had tightened by 100bps at a time instead of 25bps, credit supply would have turned around far earlier. Your belief that central banks are incapable of halting credit expansions when they are determined to stop them is literally insane.

    @Geoff: Don’t get me wrong, I’m in the endogenous money camp. Understanding the self-perpetuating nature of credit and deposit expansion is crucial to understanding the credit and trade cycle. But it’s one thing to say the central bank typically passively plays along with credit booms, and a very different, and wrong thing to say that the central bank has no choice.

  118. This Tom fellow (not Tom Brown) seems to be confused on excess reserves and required reserves. The fed controls the overnight rate in the pre-2008 period by changing the level of excess reserves. Excess reserves put downward pressure on the overnight rate when banks try to lend them out.

    If a bank needs required reserves and the Fed has not supplied them then the Fed has no choice but to supply them or it will cause a bank to be unable to meet its legal regulatory requirements at day’s end.

  119. The Volcker recession was in the early ’80s. The recession had multiple causes but what I said was that Volcker tightened excess reserve supply and put a damper on credit expansion. That’s not debatable. Go read.

  120. “Your belief that central banks are incapable of halting credit expansions when they are determined to stop them is literally insane.”

    I didn’t say the central bank has no control over inside money distribution. You are the one claiming they have absolute control over inside money distribution. I am the one claiming they only influence a part of the spread, which certainly influences inside money issuance, but does not control it.

    And yes, the central bank has no choice in providing required reserves when the banking system needs them. This is not a matter of debate. It is a simple fact. The Fed doesn’t refuse to supply required reserves to a solvent bank. It’s preposterous to claim that it would.

  121. I am well aware of the Volcker mythology. It’s the same myth that made Bernanke assume he could contain the sub-prime crisis by raising rates. Of course, that didn’t work and since it wrecks your argument you just claim he didn’t act fast or aggressive enough.

    Well, of course, if the central bank wanted to wreck the payments system it could. I’ve never said otherwise. That’s an extreme and mostly useless claim. And it has nothing to do with whether banks lend reserves or whether the central bank supplies required reserves to the banking system after loans have been made or whether the central bank controls the supply of inside money 99% of the time. You’re just moving the goal posts as it becomes clear you don’t understand everything you claim to….

    I don’t have time to debate you on this right now so I’ll just ask that you try to be a bit more respectful and maybe not so presumptuous about what you think other people actually understand. Thanks.

  122. @LVG – That’s not how it works in practice. Speaking of the situation to 2008, banks that needed reserves bid on the interbank market for them, and the central bank controls the rate by adding a lot, a little or no reserves each day or even taking some away. Your example seems to refer to a bank that can’t afford to pay the interbank rate. Such banks typically are given central bank loans but if the situation persists the central bank will condition further loans on the bank reducing or stopping lending, or the bank will be taken over by regulators.

  123. Tom,

    There’s absolutely nothing wrong with my definitions of reserves or base money. They are the common, usual definitions and are not my “own special definitions.”

    I’ve been pointing out here for a long time the three paths you sugggest:

    1. To a non-bank Fed deposit holder
    2. Cash to the private non-banks
    3. Back to the Fed

    But technically, it’s the base money that leaves. Reserves are just diminished in the process.

    The bigger picture is that regarding the large macro picture, reserves move around from bank to bank or between the banks and the Fed …. all at the well established FFR… and that continues so long as the Fed chooses to target an FFR: they have every means at their disposal to make that happen ER > 0 or not.

    Cash advances then are the sole means by which base money can be loaned out to the non-bank public.

    The important thing that I think Cullen was trying to get across is that ER > 0 or not doesn’t matter one iota: Reserves are not going to flood out into the private sector any more so now than they were prior to 2008 and for as long as the Fed has been targeting an FFR (basically the entirety of the “Great Moderation” and post GM up until now: several decades altogether). With FFR targeting (even if there’s a longer term inflation rate targeting happening simultaneously), reserves are there to facilitate inside money and nothing more. They will be provided so as to accomplish that and at the same time hit the FFR target. Excess reserves don’t matter in this regard.

  124. When the Fed announces it will stop targeting and defending an FFR, this will change, but not until then. To suggest otherwise is wrong.

  125. Tom,

    I think you missed my point. The level of loans/deposits determines required reserves endogenously. The Fed MUST supply the reserves to settle payments and meet reserve requirements. It does this at a price (the overnight rate), but it always supplies reserves when needed.

    Think of required reserves as overdrafts by banks at the Fed. The Fed has to supply the reserves to resolve an overdraft in the interbank market or else it’s causing a solvent bank to miss its obligations.

  126. Thank you. The key to understanding endogenous money is actually understanding that required reserve creation is itself endogenous (or rather, the result of endogenous money)! And excess reserves don’t result in more inside money creation via lending.

    As Krugman said, I hope we’re at a level of impressive clarity now! But my guess is Tom here will keep arguing even though he’s now clearly proven that he doesn’t understand how required reserves are endogenously the result of the loan process….

  127. Alright, we’re going in circles.

    Your logic is simply wrong. To clarify, we’re speaking of situation in the US to 2008. Banks didn’t suddenly accidentally discover at the end of some day that they had run out of excess reserves and go running to the central bank for more. Such an example has nothing to do with how credit expansion and base money expansion proceeds.

    The central bank regulates the pace of lending through controlling the supply of excess reserves, which in turn regulates interest rates, which in turn regulates demand for credit. In other words, the ultimate goal of manipulating reserves is to encourage or discourage people and companies from borrowing. That is a very powerful tool, in a situation where excess reserves are tight.

    There was nothing mythological about the interest rates that banks were asking for loans during Volcker’s clamp-down. They asked such high rates because Volcker had made excess reserves supply very tight. There are many other examples around the world of episodes when central banks determinedly clamped down very hard on credit expansion.

    I never said central banks have exclusive control over credit expansion and broad money supply. Yes, in practice, central banks are reluctant to clamp down too hard on lending and cause bank failures (as indeed Volcker arguably caused all those S&L failures, or at least brought them to a head). Central banks do indeed more often play along with or even foment the credit cycle rather than moderate it. But that’s getting into the psychology of regulators. We were discussing the capabilities. If your argument were that central banks have a long record of doing nothing or too little to slow credit bubbles, I would agree absolutely. But it’s not serious to say that the central bank can’t slow or even stop credit expansion if it really wants to.

    I’d suggest you study emerging market central banking, which is much like central banking was in the US until 2008, with rapid credit expansion and significant inflation and some central banks fomenting, others playing along and others resisting overheating. It might help you understand how things worked in the US until 2008 and how different central bankers either assert themselves in different ways. Central banks are after all very powerful. Ours has simply, sorry for the crudity, shot its wad.

  128. 1) I’ll kindly ask that you stop saying everyone else is wrong. There are now 4 people trying to explain the same point to you that you don’t understand. So it’s obviously only 1 person here who doesn’t get it.

    2) Banks don’t “run out of” excess reserves. They’re excess. This is the basics….

    3) The level of excess reserves does influence lending. But it only influences the cost of overnight lending which can alter the spread at which banks will make loans. But it does not mean banks can’t make loans nor does it directly result in fewer loans. If the banks can pass the cost along to their customers through higher rates or by reducing the profit spread then the loans will still be made regardless of what the central bank does. Again, this is basic banking.

    4) Yes, central banks support banks more than they go against them. Central banks exist primarily to support banks after all! But we agree, in unusual cases, the central bank might clamp down on lending in very aggressive or unusual ways to stomp out inside money creation. That’s very very rare though and largely inapplicable to how everyday lending works (which was the whole point of this discussion which you’ve now derailed thanks to your accusations and misunderstandings). More often, they accommodate bank lending however they can.

    5) I’d suggest you stop making suggestions since you’ve clearly proven that you’re the one who doesn’t entirely understand everything here.

  129. 1) Fifty million Elvis fans weren’t wrong but Cullen Roche and three like-thinkers definitely are wrong. Sorry, this is a factual debate. Facts are facts.

    2) Banks do run out of excess reserves, any time their reserves are less or equal to required reserves. At the individual bank level, in a large, tight reserves banking system like the US before 2008, it happens to multiple banks per day. On a system level, excess reserves in practice never fall to zero, but supply does loosen or tighten, and that influences interest rates, which influences demand for borrowing.

    3) Yes. Higher interest rates suppress demand for loans, but if for other reasons demand for credit is rising by more than higher rates are suppressing it, credit expansion can still accelerate. Still, the central bank could slow or stop the expansion if it wanted, by tightening more sharply.

    4) It’s true that central banks most often encourage lending, not discourage, but that’s not good enough to assume that mode when explaining how banking works. You’re leaving out the whole tightening cycle.

    5) Fine, don’t educate yourself and continue to believe and teach things that aren’t true. But actually I do understand you, and I laid out very plainly why you’re wrong. You seriously should study emerging market central banking.

    Central bank money expansion is neither entirely endogenous nor entirely exogenous. It tends to mostly follow an endogenous process as a rule of thumb. But it’s also an unpredictable policy variable.

    And so goodnight.

  130. 1) Sounds like a Copernican moment for you of sorts….

    2) We’re talking aggregate, but you knew that.

    3) You’re repeating what I already said….

    4) You’re assuming the tightening cycle is much more powerful than you think and repeating your belief that the Fed controls inside money via outside money.

    5) No, you came here with an argumentative attitude and then you proceeded to lecture me on how you understand everything. Then you proved that you don’t understand how new loans will at times create required reserves or overdrafts at the Fed by accounting necessity. In other words, you proved that you don’t understand one of the core components of endogenous money while lecturing everyone here how you understand everything about endogenous money….

    Good night!

  131. At the risk of pointlessly trying to get last word, I’ll respond to your claim that I misunderstand one more time. I’m not misunderstanding anything. I’m pointing out clear, incontrovertible facts, none of which you have actually contradicted.

    I never said the central bank “controls inside money”. Central banks certainly have a great deal of influence over the supply of credit and bank deposits. In the pre-2008 US situation, and current emerging market situation, central banks have a very powerful lever to either encourage or slow the growth of broad money: manipulating the supply of excess reserves. In the current US situation, the central bank can do relatively little to encourage faster credit and deposit growth, as rates are near zero. You can’t deny those facts, so stop claiming I misunderstand something.

    Likewise, I’ve already explaimned to you that your example purporting to show how central banks are forced to increase base money by as much as banks desire to lend is not a realistic example. I’ve already explained how the real-world relationship between central banks and commercial banks works in the pre-2008 US system and others like it: the central bank manipulates excess reserves supply, which in turn moves interbank rates to a certain level, which influences the demand for credit and the banks’ judgments of customer creditability and thus influences the pace of expansion. If the central bank thinks credit expansion is too quick, it raises rates.

    When banks are short of reserves as a result of the day’s operations, they don’t go first to the central bank, they go to the interbank market. The central bank only provides as much reserves as necessary to keep interbank interest rates from rising above its target. That is how the central bank limits the pace of credit expansion.

    In normal times, banks go directly to the central bank only when they are singled out by other banks and cut off from access to interbank lending at general market rates. They typically get help to meet their obligations but are discouraged from or even banned from further lending.

    If many banks come all at once to the central bank needing reserves, that’s a banking crisis and another subject, which has nothing to do with how broad money expands in a normal economy.

    Again, you can’t contradict these facts. You can’t point to a real-world example where banks in general have lent more than the central bank wanted them to and forced the central bank to allow faster credit expansion than it wanted to allow. So stop claiming I misunderstand something.

    Endogenous money is a very important issue, dear to my heart. So, excuse me, but it riles me when guys like you make it appear that only wingnuts believe in endogenous money. You’re losing the argument to Krugman by making rash absolute statements about how banking works, and you’re just wrong.

    Endogenous money doesn’t need to deny the undeniable power of central banks, or the undeniable truth that banks lend out reserves, or the undeniable truth that lending slowly consumes excess reserves. Endogenous money lives in the messy real world. Get off the MMT-ish alternate reality magic bus and get your hands dirty learning the real-world banking business.

  132. Okay, let me repeat my positions then since you haven’t absorbed them or taken the time to begin to understand them.

    1) Raising interest rates is a very blunt instrument, which, as the 2002-2007 period proves, doesn’t always work. This doesn’t mean monetary policy is useless or that I would get rid of it, but I would be more careful and precise than to refer to this as a “very powerful” tool as you have in these comments. Again, lending is primarily a DEMAND SIDE issue that you’re trying to frame as a supply side issue. Solvent banks lend when creditworthy customers have demand for loans. Not when the Fed decides they want loans. The counterexample of the weakness of raising rates is the current environment where lowering rates hasn’t worked because demand for debt isn’t very high. Yet, somehow, you’ve miraculously ignored this demand side effect! The Fed only impacts this relationship indirectly by influencing the spread at which banks make a profit. You claim it has to do with “absorbing” excess reserves which is just a total misunderstanding of the main driver of loan accumulation and its relationship to reserves. You have the entire understanding precisely backwards! And yes, you are misunderstanding this basic point.

    2) I was not creating an “example” about how a central bank must provide required reserves if they don’t exist. It is an irrefutable fact of central banking. If a bank makes a loan that causes a deficiency in required reserves then the equivalent of an overdraft occurs at the central bank. The central bank can either choose to disrupt the payments system by not supplying the reserves or it can issue them. The thing you don’t understand is that this process happens automatically. The central bank has no choice in the matter. Banks essentially force the central bank to supply reserves to meet reserve requirements. Again, you have the understanding precisely backwards.

    3) I don’t deny that central banks have extraordinary power. I am not against monetary policy. So I don’t know where you got that notion from. It just seems to be one of the many misunderstandings you have after having introduced yourself to this website 3 days ago when you jumped to the ridiculous conclusion that you understand everything I’ve ever written….

    4) I am not an MMTer and if you actually knew the first thing about me you’d know that I vehemently reject MMT. But again, you don’t know me or my positions yet you come here beating your chest leaving these absurdly arrogant comments as if you do know me. What is this now? The 4th or 5th thing you have dead wrong in just ONE paragraph? Do I need to keep going?

    Sorry, but your comments have been arrogant and rude since the very start and there’s very little of any use that’s coming from this. I wish you’d been a bit less combative when you decide to start commenting here because you seem like you probably agree with me on quite a bit. Unfortunately, you jumped to lots of conclusions and decided to hide behind anonymity as you lept into attack mode. You’ve refused to try to understand how you misunderstand these points so I will just ask you to kindly take your know it all attitude back to Krugman’s site where you can protect his liquidity trap view of the world. Thanks.

  133. Tom, so you’re basically saying that prior to 2008, over long time frames, the Fed did inflation targeting by adjusting (every six weeks or so) its fixed target FFR by using something like a Taylor rule. I don’t think anybody here disputes that!

    And in between those six week Fed meetings, it provided (primarily through repos) exactly the reserves required at all times to achieve the fixed FFR target.

  134. Tom, again all you basically said was “The Fed used to use (pre-2008) a Taylor rule”… but on top of that you added statements like this:

    “You can’t point to a real-world example where banks in general have lent more than the central bank wanted them to and forced the central bank to allow faster credit expansion than it wanted to allow.”

    To make it sound like the Fed closely monitors “credit expansion” on a minute by minute basis… as if “credit expansion” was their overriding concern! They are certainly FREE to make it their overriding hourly concern, but the fact is it IS NOT, nor has it been: On a daily/hourly basis they REACT to what the banks do: they honor overdrafts and add or remove reserves through repos to hit the FFR. That’s their daily, hourly concern! Not “credit expansion.” Of COURSE they don’t set out to enable nearly insolvent banks by lending to them through the discount window. Those are the exceptional cases, not the typical ones. Every six weeks or so, they adjusted the FFR target in line with the Taylor rule so that over a two year time frame or so they can track their inflation target.

    So on a short term basis their concern is the FFR. On a long term basis it’s the inflation target. Over no time interval do they micro-manage “credit expansion.”

  135. Alright, I’ll quit your site.

    But:

    I never said the growth of base money, broad money or credit are solely determined by the supply side. Demand and supply are both important. You’re the one with the odd and unsustainable position that only demand matters.

    If your point were that inflation targeting won’t stop credit bubbles, I would agree that the 2002-2007 case proves that. But it proves nothing about central bank’s ability or inability to stop credit expansions. The central bank has absolute power to stop credit expansions. All you need to do here is stop overstating the case. Yes, central banks usually let the credit cycle run. But that’s a policy choice, not a given.

    What you wrote about central banks automatically providing funds to banks that “overdraft” their reserve accounts is not only refutable, it’s kooky. That is not what happens. What happens is banks end the day short of required reserves and must borrow on the interbank market to cover the gap. The central bank *absolutely does not* automatically issue reserves. If you don’t believe me you’ll just have to ask any banker. Where did you get that misinformation? Anyway, it’s wrong. Leaving banks to solve their own reserve shortfalls on the interbank market does not cause any disruption to the payments system. It’s a crucial part of how the central bank rate-targeting mechanism works.

    Only if the bank can’t or won’t resolve its required reserve shortfall on the interbank market does the central bank provide funds to what is then considered an ailing bank. Such last resort lending is insignificant to a credit expansion. The Fed and most central banks use sovereign bond purchases to supply reserves to credit expansions. I’ve never heard of a bank overdrafting its reserve account. In the US the FDIC would take over before then.

    The amount supplied is consistent with the interest rate target, and it is the rates that restrain lending, as I’ve explained. I have not put anything backwards. I’m just explaining how the supply works. I’m not saying supply is the be-all and end-all. Supply and demand are always important.

    I’m sorry if it insults you to be compared to MMT-ers, but you do claim that banks don’t lend out reserves, which is a big part of MMT. Are you sure you didn’t get the misinformation about banks overdrafting reserve accounts and central banks automatically replenishing from an MMT text? Sure sounds like something they would come up with. Anyway, I did say MMT-ish.

    I’m not at all a Krugmanite. I don’t use the term “liquidity trap”. I have criticized his IS-LM model, which makes no sense (what can it mean to show a model in which savings are greater than investment, when savings are by definition less than investment by the current account deficit?) Krugman is of course right that when rates are near zero, increasing the money supply doesn’t lower rates further, but duh. I think near-zero rates describes the situation sufficiently.

    Okay, that’s it. I’m sorry you think you got nothing out of it. And I’m sorry I sound arrogant, but I do know what I’m talking about. What can I say, hear me now, believe me later. Go find where you got the overdraft and automatic replenishment misinformation, bring it to a banker, and let him have a laugh at your expense. Then reflect on what nonsense you’re capable of making yourself ardently believe, and remember me. Goodbye.

  136. Tom,

    Overdrafts aren’t a myth. They happen very regularly and the FDIC doesn’t seize the bank because of it (that’s absurd). And you can track them. See this document from the Fed proving you wrong:

    http://www.federalreserve.gov/paymentsystems/psr_dlod.htm

    The only thing “kooky” here is that you’re not even aware of the basics, but you’re coming here writing some of the most arrogant stuff I’ve ever read. How can you possibly claim to be an expert on banking when you aren’t even aware of the overdraft facilities at the Fed?

    Of course, this isn’t occurring in this environment or really ever because of reserve requirements, but that’s not the point. The fact is, if the banking system as a whole has solvent banks who can’t meet Fed regulations then an overdraft is provided automatically. You can google “daylight overdraft federal reserve” to read all about this new thing you just discovered. It will round out the “fact” that you already know everything about banking! :-)

    Also, I never said supply doesn’t matter. But it’s usually not the driver of loan creation.

    Anyway, you latched onto a bunch of inapplicable and rather meaningless points and then ran to timbuktu with them so we’ve really just wasted a whole lot of time here. And in the process your arrogance backfired on you when your exposed the fact that you don’t really know nearly as much as you claimed….Maybe try to be a bit more cordial next time rather than running around anonymously beating your chest in people’s faces and calling them wrong in every comment. Just a thought.

    Cullen

  137. Alright, hi again Cullen.

    Well, congratulations, you were partly correct on one small point and I was wrong.

    Small US banks do regularly overdraft their reserve accounts. These are typically very small banks whose transaction deposits are less or not much more than the barrier between the 3% reserves ratio levied on small amounts of transaction deposits and the 10% reserves ratio levied on most transaction deposits: http://www.federalreserve.gov/monetarypolicy/reservereq.htm#table1

    So I stand corrected on that point. The smallest US banks have such small required reserves that they can run their reserve accounts below zero even when they are healthy. I didn’t know that. My view that with a 10% reserve ratio on deposits any bank with less than zero reserves must be failing only applies to mid-size and larger banks.

    But, you are mischaracterizing these overdrafts as something important to the process by which demand for credit leads to increases in reserves. They’re not. These are very short-term overdraft loans. Any bank that takes them, like any bank that falls below its required reserves, still must raise funds on the interbank market. The Fed supplies reserves through asset purchases, as much as is required to keep the target. Banks can’t force the central bank to permanently increase or even significantly increase the reserves supply through these overdrafts.

    The overall case remains as I said, in a regime where the central bank keeps reserve supply tight and targets the interbank rate, the central bank is limiting lending by limiting the supply of reserves to the amount needed to support the amount of borrowing that banks and borrowers can agree to at such rate levels plus usual spreads. Both the supply of reserves and the demand for credit are important to determining the amount of credit and deposit expansion that will happen. At higher interest rates, lending and deposits will expand less than they would with lower rates. It’s silly to claim the central is forced to allow however much credit the market demands. The central bank can always suppress demand by raising rates.

    One reason I pointed you to emerging markets is that you can see there how this works in practice, which you can’t do anymore in the US, as it has changed regimes. You can also see central banks such as China and Turkey that informally target credit expansion rates (not at all strictly, and among other targets that can’t practically be strictly targeted simultaneously). It is definitely within a central bank’s power to target a growth rate of credit, deposits or broad money.

  138. You should have just stopped after saying “you were correct” because I never said half the other stuff the rest of your comment is about….

    Take care.