One general theory of money creation to rule them all!

The Bank of England has published a lovely clear article (by Michael McLeay, Amar Radia and Ryland Thomas) on "Money Creation in the Modern Economy". Thanks to JKH for the tip-off. (Here is JKH's blog post). But I disagree with it.

Thinking about monetary policy in terms of interest rate policy just doesn't work. It doesn't work in theory, and it doesn't work in practice. That's why the Bank of England is having to do QE, and is having to re-introduce the old general theory of monetary policy as a special theory for QE.

This is the part of the article that most caught my attention:

"Like reductions in Bank Rate, asset purchases are a way in which the MPC can loosen the stance of monetary policy in order to stimulate economic activity and meet its inflation target. But the role of money in the two policies is not the same.

QE involves a shift in the focus of monetary policy to the quantity of money: the central bank purchases a quantity of assets, financed by the creation of broad money and a corresponding increase in the amount of central bank reserves.
The sellers of the assets will be left holding the newly created deposits in place of government bonds. They will be likely to be holding more money than they would like, relative to other assets that they wish to hold. They will therefore want to
rebalance their portfolios, for example by using the new deposits to buy higher-yielding assets such as bonds and shares issued by companies — leading to the ‘hot potato’ effect discussed earlier. This will raise the value of those assets and lower the cost to companies of raising funds in these markets. That, in turn, should lead to higher spending in the economy."

There is nothing wrong, in my eyes, with that second paragraph. Good monetarist hot potato stuff! What is wrong is the sentence that immediately precedes it: "But the role of money in the two policies is not the same."

They have two theories of how monetary policy works. There is one theory for when the Bank of England sets a rate of interest: "And reserves are, in normal times, supplied ‘on demand’ by the Bank of England to commercial banks in
exchange for other assets on their balance sheets." And a second, quite different theory, for when the Bank of England does QE.

I'm sure they are not alone in having two theories: one for "normal times"; and one for QE, which is seen as needing a special theory only applicable in "abnormal times". But it is rather peculiar having two different theories of the same thing.

One theory is better than two.

What is even more peculiar is that their special theory for QE is the same as the general theory taught in first-year textbooks. The central bank buys a bond and the money supply expands, because the seller of the bond now owns the money that the central bank gave him in exchange for the bond.

"Quantitative Easing" is just a silly new name for the "Open Market Operations" that first-year textbooks have always said was the way that central banks normally increase the money supply.

If you spend your life teaching first year economics, like I do nowadays (when I'm not blogging), this here modern world looks very peculiar. My general theory has become their special theory, and they have gone and invented some weird new theory of money creation in what they call "normal times", that they admit doesn't work as a general theory, and they still need my old theory to handle the cases where their weird new theory doesn't work.

Here is my general theory: when the central bank buys something, with central bank money, the money supply expands, because whoever sold them that something now holds extra money. Done. It does not matter whether the central bank buys a bond, or a computer, or whatever. Hell, it could just give the money away to its favourite charity (helicopter money), and the result would be the same.

Why can't my general theory work equally well in "normal times"? Let me repeat that above quote, this time with bold added: "And reserves are, in normal times, supplied ‘on demand’ by the Bank of England to commercial banks in exchange for other assets on their balance sheets." See that bit about "in exchange for other assets"? That means the Bank of England buys something. Just like I said in my general theory. The central bank increases the money supply by buying something. See, it's easy!

Now if the central bank is buying something, and someone else is selling something, there must be some sort of market in which that something is bought and sold. But it really doesn't matter, for money creation, what that "something" is. What matters is that the central bank is selling central bank money. It is supplying central bank money. So we want to know something about the central bank's supply function. And that supply function will depend on what it is the central bank is targeting.

We could assume that the central bank is targeting the stock of its own money, so the supply function is perfectly inelastic with respect to anything. That is a very simple assumption, suitable for a first year textbook. But not very realistic, for most times and places.

We could assume that the central bank is targeting the price of gold, so the supply function is perfectly negatively elastic with respect to the price of gold. Realistic in the past, but not nowadays.

We could assume that the central bank is targeting the stock of M1, so the supply function is perfectly negatively elastic with respect to the stock of M1. Realistic briefly in the past, for Canada, but not nowadays.

We could assume that the central bank is targeting expected CPI inflation, so the supply function is perfectly negatively elastic with respect to the expected rate of change of the price of the CPI basket of goods. That is realistic for many central banks nowadays.

Interest rates? Did I hear you say that modern central banks target interest rates? Well, maybe, but only in the very short run, like maybe the next 8 weeks. Monetary policy for the next 8 weeks isn't very interesting, unless it gives us a clue about what the central bank will be targeting in the years after that. And they say they are targeting things like inflation, not interest rates. But if you insist, the central bank's supply function would be perfectly elastic with respect to whatever interest rate you say it is targeting.

See, it's easy. One general theory to rule them all, that can be modified for whatever it is you want to assume the central bank is targeting, just by changing the central bank's supply function.

But what about commercial banks? They create money too. Commercial banks are much easier. We know what commercial banks are targeting. They target maximum profits. And that means commercial banks, like ordinary profit-maximising firms, and like ordinary utility-maximising people, and unlike central banks, only care about real variables. It is the central bank's job to ensure that nominal variables are determinate, by not doing something daft like trying to target a rate of interest for too long.

Just like central banks, commercial banks create money by buying something, and paying for it with the money they create by buying it. They mostly buy non-monetary IOUs, but it doesn't matter what they buy, or even if they just give their money away to their favourite charity.

How much money commercial banks create to maximise their profits will depend on a lot of things. But what I want to focus on, because this is the key policy question, is how it depends on what the central bank is doing. Let me quote again from the Bank of England article:

"The supply of both reserves and currency (which together make up base money) is determined by banks’ demand for reserves both for the settlement of payments and to meet demand for currency from their customers — demand that the central bank typically accommodates.
This demand for base money is therefore more likely to be a consequence rather than a cause of banks making loans and creating broad money."

First, they don't mean "supply"; they mean "quantity supplied". And as I pedantically tell my first year students, the difference really does matter (sometimes, like here). Yes, the quantity supplied (which is equal to quantity demanded in equilibrium) depends on the demand function, but it depends on the supply function too. Both blades of the Marshallian scissors matter in determining quantity, even if one blade is assumed to be horizontal for the next 8 weeks.

But what matters is their acknowledgement that the demand for base money (central bank money) is a consequence of the amount of broad money commercial banks have created. May I make a small simplifying assumption? May I assume that the demand for base money is proportional to the stock of broad money, other (real) things equal? Because that's the only way we can assume that commercial banks maximise profits and so only care about real things and don't suffer from money illusion. Thanks!

"While the money multiplier theory can be a useful way of introducing money and banking in economic textbooks, it is not an accurate description of how money is created in reality. Rather than controlling the quantity of reserves, central banks today typically implement monetary policy by setting the price of reserves — that is, interest rates."

But hang on! You have just agreed (sort of) that the demand for base money is some proportion r of the stock of broad money. So in equilibrium, when the actual stock of base money is equal to the quantity of base money demanded, the stock of broad money must be a multiple 1/r of the stock of base money. And if the central bank shifts the supply function of base money $1 to the right, that must increase the equilibrium stock of broad money by $(1/r). Just like the first-year textbook says it will!

Now you might object that modern central banks don't care about the stock of base money (except when they are doing QE), and target things like inflation instead (except for 8 week periods when they target an interest rate). OK. But if the central bank wanted a temporary increase in the inflation rate, and so a permanent rise in the price level, it would need to shift the supply function of base money, to create a permanent rise in the monetary base, and a permanent rise in broad money, and the textbook money multiplier would tell us that broad money would increase by 1/r times the increase in base money.

One simple (first-year textbook) general theory to rule them all!

What is the underlying problem here? Why do monetary economists resist this very simple and very general theory of monetary policy? The underlying problem is revealed in this quote:

"Like reductions in Bank Rate, asset purchases are a way in which the MPC can loosen the stance of monetary policy in order to stimulate economic activity and meet its inflation target."

It assumes that interest rates are a measure of the "stance of monetary policy". If interest rates were an adequate measure of the "stance of monetary policy", the Bank of England would not need QE. And you cannot define the stance of monetary policy by taking some sort of average of interest rates and QE. A permanent increase in the target price level would mean a permanent increase in the money base but would have no obvious implications for interest rates. A permanent increase in the inflation target would mean a permanent increase in the growth rate of the money base but would mean higher nominal interest rates. There is no monotonic mapping from loose monetary policy into low interest rates. Thinking about monetary policy in terms of interest rate policy just doesn't work. It doesn't work in theory, and it doesn't work in practice. That's why the Bank of England is doing QE, and having to re-introduce the old general theory as a special theory of how monetary policy works.

Comments

[Scott Sumner]

I wrote a post on this yesterday, but haven't put it up yet. Just as well, yours is better.

[Nick Rowe]

Thanks Scott! (I feel slightly guilty for stealing your idea that commercial banks only care about real things. It's implicit in Patinkin, I think, but you made it clear and explicit for me.)

You should put your post up now. See where we overlap and where we don't.

[Tom Brown]

Nick, thanks for going over that. You didn't write anything which surprised me... I must be starting to learn your views.

[Marko]

I have more bad news , guys.

There is no Santa Claus.

Man , reality sucks.

[Nick Rowe]

Tom: yep. You can usually guess what I'm going to say. Which proves you are learning stuff (whether you agree with me or not).

Marko: you lost me.

[alex]

Nick, Why do you write such lengthy posts? Could you put only significant parts here?

[Jason]

Thanks for the link and the discussion, Nick. A lot of great information that really helps me think about this stuff.

I found a function that describes the broad trend of US interest rates in terms of the monetary base (graph at link)

http://informationtransfereconomics.blogspot.com/2014/02/the-link-between-monetary-base-and.html

The fit for both the short run interest rate and the long run rate uses the function c log r = log (1/κ) (NGDP/Mx) with κ = 10.4 and c = 2.8, and Mx being either the currency component ("M0", for the long run rate, 10-year) or currency + reserves (MB, for the short run rate, 3-month). You can plot it yourself :)

One thing that this indicates is that the level of interest rates is never indicative of "tight" or "loose" money -- the relevant measure is the rate of NGDP growth relative to base growth.

PS: I answered your question on a previous post with new post ... http://informationtransfereconomics.blogspot.com/2014/03/apples-bananas-and-information-transfer.html

[Tom Brown]

Nick, quick hypothetical question (I know how you love your hypotheticals!). Here's the SUPER simple setup: A CB and one commercial bank. No cash, taxes, government spending, or foreign trade. It seems to me that in such a circumstance base money (consisting only of reserves in this case) might well be $0 (unless the CB does OMOs), after all there's no need of reserves to interface with the gov or to settle transactions, or to be withdrawn as cash. But that doesn't necessarily mean that M1 is $0, does it? I realize you say that a ratio between the two is a reasonable approximation, but I don't see how a finite ratio applies in this case.

So if MB = 0 then VB goes to infinity doesn't it? (Assuming NGDP > 0).

If that's the case, what does the one theory to rule them all say about how prices should change in the long term if the CB now buys $X in assets?

[The Market Fiscalist]

It would be both true and elegant to say something like "there is a positive correlation between the money supply and NGDP. No matter what means you use to increase the money supply it will (other things equal) always boost NGDP"

It would also be true (but perhaps less elegant) to say: "There are a number of different mechanism by which an increase in the money supply can be used to increase NGDP. Depending upon the mechanism used (govt deficits v CB asset purchases) and other variables in the economy (the degree of prices stickiness and the current nominal interest rates) the increase on the money supply will affect NGDP in different ways and to different degrees and have different secondary effects."

Perhaps you could argue that there is some underlying "pure" monetary effect and everything else is just distributional - but that seems to be something of an oversimplification when you look at the difference between an NGDP increase due 1) to a taxation cut 2) to a lowering of the target interest rate and 3) to the purchase of a high-risk MBS as part of a QE program.

[Tom Brown]

... given my simply hypothetical, it still seems the commercial bank has an opportunity to make money for its shareholders by lending money, no? I also just don't see any demand for reserves. What am I missing? I can see it acquiring a nice amount of equity with which it credits the deposits of its shareholders, making them rich and happy. You might ask "What is the CB targeting?" Suppose they were targeting base money stock at $0. Then you can assume they switched to a target of $X.

[Nick Rowe]

alex: because everything I wrote there was significant! Plus, my last post was very short, and I wanted to keep a balance!

Jason: "One thing that this indicates is that the level of interest rates is never indicative of "tight" or "loose" money -- the relevant measure is the rate of NGDP growth relative to base growth."

That sounds sorta good to me (though I would say that NGDP growth relative to base growth might be a better measure of *turning points* in monetary loosening), but I didn't really get how you came up with it.

TMF: I could go with both. Helicopter money would be the "pure" case, and everything else would be in the "other things equal that might not be equal in practice" clause.

[Tom Brown]

I got two in spam.

[The Market Fiscalist]

Nick, OT but I was interested in your comment 'Helicopter money would be the "pure" case'. At the ZLB when you have a choice between creating new money via QE or via govt-deficits - isn't govt deficits (funded by things like income tax cuts) more like the "pure case" than QE ?

[JKH]

Nick,

Many points, some rough comments (at risk of haste):

The BOE paper:

"And reserves are, in normal times, supplied ‘on demand’ by the Bank of England to commercial banks in exchange for other assets on their balance sheets."

If that means demand from the private sector (I think it must), that’s wrong. (As I noted in my post and comments, the paper certainly isn’t perfect.) The BOE holds the option to transact – not the banks. And the CB enters the market on the basis of interest rate conditions – be that short term positive interest rate targeting, or QE zero boundedness issues. Even though you assert that it’s all about money in the end, I don’t think you can deny that both OMO and QE as CB strategies can be related to interest rates in this way.

(LLR is different, but mostly we’re looking at OMO and QE here.)

I think your general theory applies to currency (banknotes) in all environments – non-QE and QE.

Central banks are short a call option with regard to currency – they must supply the quantity demanded. (By that I mean an option to transact – not a fixed option strike price)

Conversely, the private sector is long a call option to acquire currency. They just pay with deposits/reserves.

The difference insofar as your general theory is concerned is in the application to bank reserve balances, IMO.

(Because of this difference, IMO, I think the monetary base definition, which conflates reserve balances with banknotes, is quite a perilous conceptual construction for monetary theory. I think that underlying everything, this may be at the heart of the disagreement between the way you guys look at the world and some of the “new” stuff.)

So in the case of bank reserve balances:

Central banks are long a transaction option with regard to both regular OMO and QE bond purchases – they have a CHOICE as to whether to transact.

Conversely, the private sector is NOT long an option to sell bonds in this way. They must wait for an offer of quantity demanded from the central bank (again, referring to an option to transact, not a strike price for an option).

That option structure difference between currency and reserve balances is key to supply demand economics, I believe.

Maybe more on that later.

A different but connected point:

“May I assume that the demand for base money is proportional to the stock of broad money, other (real) things equal? Because that's the only way we can assume that commercial banks maximise profits and so only care about real things and don't suffer from money illusion.”

You can assume that for currency (banknotes) held by non-banks, but you can’t assume that for bank reserve balances.

Maybe more on that later.

Another point - required statutory reserves must be ignored in the economics. Because some reserve regimes have no requirement, the focus must be on excess reserves as the guide to the supply demand economics. Otherwise, a zero required reserve regime will break whatever model is envisaged.

“But hang on! You have just agreed (sort of) that the demand for base money is some proportion r of the stock of broad money.”

I think that's your assumption. And it’s incorrect in the case of excess reserves.

The zero bound is the bifurcation point for what you’ve referred to as the “new” general theory and special theories. Given the interest rate constraint of the zero bound, I see nothing wrong with bifurcation as part of a complete model where interest rates are declared to be important.

[Nick Rowe]

JKH: "Another point - required statutory reserves must be ignored in the economics. Because some reserve regimes have no requirement, the focus must be on excess reserves as the guide to the supply demand economics. Otherwise, a zero required reserve regime will break whatever model is envisaged."

I agree that a general theory should be able to handle both regimes. I don't see that as a problem. Required reserves, if they exist, are one of many things that affect desired reserves. But we should abolish the (US?) term "excess reserves", because it only makes sense in a regime with required reserves. What matters is desired reserves, and the difference between actual reserves and desired reserves (I would like to redefine "excess reserves" as the difference between actual and desired, not actual and required, but doing so would probably cause confusion). What matters is whether desired reserves are proportionate to broad money, holding real things constant. If they are proportionate (and if desired stock of currency is proportionate too), we get the money multiplier in equilibrium.

[JKH]

"But we should abolish the (US?) term "excess reserves", because it only makes sense in a regime with required reserves. What matters is desired reserves, and the difference between actual reserves and desired reserves (I would like to redefine "excess reserves" as the difference between actual and desired, not actual and required, but doing so would probably cause confusion)"

On the first part, required reserves = 0, so "excess reserves" still works in that context, IMO.

But I hear you on the second part. OMOs are really about the CB tweaking the quantity of excess reserves at the margin - which goes to your point.

I started a draft post about a year ago playing around with that terminology. I had something but will have to look it up. The point is relevant as you frame it and is quite relatable at the operational level.

[Nick Rowe]

JKH: "That option structure difference between currency and reserve balances is key to supply demand economics, I believe."

Interesting. I'm not sure I understand it yet.

Would I be correct in saying that the central bank must swap reserves and currency at one-for-one either way as the commercial banks wish? (Otherwise $1 reserves won't be worth $1 currency). But it can make the total of the two whatever it wants, by buying or selling something. That's how I think of it.

[JKH]

Sorry, my 5:02 - I'm good up until your final sentence there - separate issue of institutional differences as in my first comment, I think.

[JKH]

“Would I be correct in saying that the central bank must swap reserves and currency at one-for-one either way as the commercial banks wish? (Otherwise $1 reserves won't be worth $1 currency). But it can make the total of the two whatever it wants, by buying or selling something. That's how I think of it.”

I think that’s very generally right – that would be the net capability of the CB holding the option to transact in OMO/QE - taking into account the “option” exercised against it on currency, and taking into account a whole bunch of other stuff.

But in saying “whatever it wants”, you’re really implying that QE is always an option in response. And I know you’re not differentiating between OMO and QE, but QE requires a special constraint in terms of the requirement to pay interest on excess reserves – if the supply of excess reserves moves beyond the inelastic portion of the demand curve for reserves in the case where no interest is paid. That’s the way I think about the transition from the pre-2008 Fed to the QE Fed and IOR.

[ATR]

"OMOs are really about the CB tweaking the quantity of excess reserves at the margin - which goes to your point."

I don't see why the word 'excess' needs to be used, as opposed to just reserves. Pretend there's only one big bank to simplify. The CB can also supply a quantity of reserves under the required level, whether that's some proportion of demand deposits or 0. Unless you want to call that 'negative excess reserves.' It just depends on the interest rate they want to hit within their corridor.

There's a 3-step timeline, assuming a 1 day reserve period. The CB sets the quantity of reserves to be traded in the interbank market. The interbank session is held and closed. The reserve maintenance period ends at the end of the day and banks' reserve positions are evaluated. The uncertainty of liquidity flows between the last 2 steps is what gives rise to the elasticity in the demand curve. It also allows the CB to set quantity of reserves above or below the requirement to achieve various rates. So whether the CB aims to supply excess or deficient reserves depends on the rate they want to hit.

[JKH]

"So whether the CB aims to supply excess or deficient reserves depends on the rate they want to hit."

That's right.

I mean excess reserves as a category - whether the quantity setting is positive or negative.

[ATR]

Okay got it.

[Nick Rowe]

JKH: "And I know you’re not differentiating between OMO and QE, but QE requires a special constraint in terms of the requirement to pay interest on excess reserves – if the supply of excess reserves moves beyond the inelastic portion of the demand curve for reserves in the case where no interest is paid."

I still see no reason to have a special name for OMO. (Actually, I would prefer to scrap "OMO" as well, and just talk about buying stuff and selling stuff.) Paying interest on reserves is just a way to increase the demand for reserves, rather like required reserves.

[Tom Brown]

Nick, I've got some in spam I think.

[Jason]

Nick, you said:

"...(though I would say that NGDP growth relative to base growth might be a better measure of *turning points* in monetary loosening), but I didn't really get how you came up with it."

It comes from the model being a function of the ratio of NGDP/MB. If MB and NGDP grow at the same rate (call this "loose money" for arguments sake), then interest rates are constant at say r = r0, but the level of r0 is dependent on the history of relative NGDP to MB growth, so any given r0 is consistent with "loose money".

The ratio itself comes from assuming homogeneity of degree zero (long run neutrality of money) per Bennett McCallum's characterization of the QTM ... at least that's the less heterodox derivation; I came to a similar conclusion via information theory :)

[Nick Rowe]

Tom: I fished you out of spam.

In your (interesting) hypothetical, what would the commercial bank promise to convert its demand deposits into? If there is no currency, and no reserves, I think that promise would become empty. The central bank disappears (its balance sheet has $0 on the liability side), and the one commercial bank is now the new (privately-owned) central bank.

[Mike Sproul]

Nick:

"It does not matter whether the central bank buys a bond, or a computer, or whatever. Hell, it could just give the money away to its favourite charity (helicopter money), and the result would be the same."

OK, I see you've returned to the dark side, and forever will it dominate your destiny. So I'll address myself to any young padawans who might still be saved.

There is another rule that really does rule them all: the real bills doctrine. It was developed by practical bankers over centuries of experience, and it was the guiding principle of banking long before the academic scribblers came along with their talk of hot potatoes.

The real bills doctrine says that money should be issued in exchange for short-term real bills of adequate value. The "adequate value" clause assures that the money-issuer will remain solvent, and that every new dollar issued will be backed by at least a dollar's worth of assets newly acquired by the money-issuer. The "short term" clause protects against maturity mis-matching, and assures that even if a bank exhausts its reserves, its customers can access their money within (say) 60 days. The "real bills" clause assures that money-issuers will provide an elastic currency, which is to say that they will provide more money when farms and factories are busy and need the money, and they will provide less money during slack times when the money is not needed.

[Richard W]

Hi Nick. Long time no speak... and apologies in advance for rambling but it's been a while since I've got my thoughts down like this. I don't want to get into the details of exactly what the BoE said, and tbh I've only skimmed it, but rather what I see as the key tension between the monetarist framework and the more post-Keynesian framework and how I've come to think about them.

OK.

Take a pure quantity-managed interest rate targeting system, like the Fed before IOR days. There's just open market operations, as you say. Any "QE" that was not an "OMO" would require an offsetting OMO in order to keep reserves at the level that meets the short-term interest rate target. So far I'm with you.

But if you have IOR (or a fixed rate full allotment reverse repo facility) then you can set the short term interest rate quite independently of the monetary base. The Fed will raise interest rates long before it unwinds its asset purchases and decreases M0. I suppose you could conceptualise this as "raising the demand for the monetary base" i.e. decreasing V, and in a sense, that is exactly what it is doing - it is using IOR/FRFARRF to change the demand for the monetary base so that whatever level the base is at the reserve market clears at the rate target. But I've come to the view that this isn't really a very helpful way of looking at it. I have (2011-vintage Richard cringes here) come round to thinking that insofar as (in a world with excess reserves*) central banks can implement a short-term IR independent of the level of the monetary base, monetary policy "really is" about a path of interest rates - a path that is of course conditionalised on expectations etc.

So going back to the QE/OMO distinction, QE is an asset purchase that a) isn't required to maintain the interest rate target and b) does not require an offsetting action because the target is not quantity managed. Furthermore, in a system that is not purely quantity managed, increasing the monetary base when rates are already at zero also is a pretty terrible commitment device for keeping future policy easier. Indeed, in the 2008 minutes Bernanke called the Fed's quantity-management interest rate targeting "a relic", and discussed making IOR the official policy rate. And it's going to happen (although whether it's the IOR rate or the FRFARRF rate or both I'm not sure).

To sum up: This is all a way of saying we're now living in a world where excess reserves* are the norm and this has changed the way I think about monetary policy. Furthermore, lots of central banks have lots of different policy rates that all set short-term interest rates in a different way (and affect demand for reserves in different ways). I think we focus too much on the way the Fed does (or did) monetary policy, as opposed to the many ways it could be done, and has been done, and will be done in future.

Debating the merits of the "base money" frame versus the "short rates" frame makes sense - I think - when you're talking about a purely quantity managed pre-2008 Fed system. As you say (and I have said before too!), 8-weeks is not a macroeconomically interesting period of time. But the world we live in now is - I think - different. Monetary policy isn't about the quantity of money in a world with excess* reserves, and short rates are set by other means.

[/ramble]

*where there are no required reserves, "excess reserves" means an excess of the reserve levels that would prevail if you were purely quantity-managing the short-term interest rate target, i.e. no IOR/FRFARRF

[Nick Rowe]

Mike: suppose I were in charge of the Bank of Canada, and you had convinced me that i really really must ensure that the Bank of Canada has plenty of assets to back its monetary liabilities, and that nothing else mattered. What inflation rate would I target? I would target the profit-maximising rate of inflation. That would maximise the net worth of the Bank of Canada, and so leave plenty of capital reserves just in case anything went wrong with any of my backing assets.

[Vaidas Urba]

Nick,

I guess they have one general theory that is based on interest rate reaction functions. That theory is equivalent to your general theory, but is specified in a different coordinate space.
And they have a second theory about credit interventions which is less important to them. And this second theory is equivalent to your theory about what happens when you do a temporary increase in money supply that is subsequently reversed (i.e. not much happens).

[Tom Brown]

Nick, thanks. I did say there's no currency, but there COULD be reserves. I'm just wondering if there's any demand for them. The CB could force reserves into existence, or the bank could borrow them from the CB (I don't know why it would though). If the CB purchased assets it would directly change the reserve level, and perhaps change the deposit level too, right? Because the bank itself may not sell it the assets it's buying directly... the non-bank private sector might be the net sellers for some of those assets, and thus bank deposits would grow accordingly. It's a little indirect, but the CB could thus target the quantity of M1.

So doesn't that free the central bank up to focus on affecting things with it's OMOs while the commercial bank is free to focus on maximizing profits?

Doe the situation change if we break the single commercial bank up into multiple banks? Does the number of banks, the timing or the settlement procedure matter any in that case? In my tidy example, every penny a bank comes up short will be matched by another penny another bank will have in "excess" ... meaning that reserve balances can be zero overnight... or between settlement periods, whenever that is, provided banks with a surplus will lend every penny of surplus out.

Maybe my hypothetical is just too neatly sealed up. Maybe it's the messiness of actually wanting to have a bit more reserves than what's required (due to uncertainties) that gives the reserves their power. I'm assuming reserves are sufficient and the CB can do what it needs to w/o cash.

It seems like there's a continuum though: like we could take tiny baby steps from the real world towards my hypothetical. Would a change come all at once with one crucial baby step, or would it come gradually, and what would that look like? And which baby step would it be? What would be changing exactly? The "constant" relating base money to broad money? What is that constant a function of?

[Tom Brown]

Mike Sproul, I was thinking of you when I read that helicopter line from Nick! Nice to see you zeroed in on it.

[Nick Rowe]

Richard: Welcome back!

We seem to have been moving in opposite directions. By around 2000 I had come around to the official Bank of Canada view, that M was just a barbarous relic, with at best some minor role as an indicator with some sort of information-content (though the sign was often negative in my regressions), and with no causal role. Ultimately, of course, the Bank of Canada needed some sort of balance sheet to control anything, but that could be ignored for all practical macro questions. The only realities were the inflation target, the interest rate instrument, and a host of possible indicator variables. Nobody believed in M, except as a simple way to teach undergraduates so you got a downward-sloping AD curve, to remind us of the indeterminacy question.

The Bank of Canada was way ahead in moving to the "new" way of doing and thinking about monetary policy.

And then came 2008/9, and when interest rates failed, and the BoC brought out QE, it was like watching atheists at prayer.

And since then I've been going back to the Old Religion!

[Nick Rowe]

Vaidas: but an interest rate reaction function can never be a general theory of central banking. For example, a 100% gold reserve central bank, that just buys and sells gold.

Tom: if there is only one commercial bank, and nobody ever wants to hold central bank currency, because there isn't any, that commercial bank would never need reserves (unless it were required by law to hold them).

[Mike Sproul]

Nick:

If you were using the real bills doctrine then you wouldn't target the rate of inflation at all. You would be in competition with other bankers, and you'd offer to pay your customers whatever interest rate that the market determined. That means that when you issue checking account dollars you might pay them 2%, and when you issue paper dollars you might pay them -1% interest, meaning 1% inflation.

It's not just adequate assets that matter. You'd also have to watch out for maturity mismatching, and you'd have to provide an elastic currency. Both of these things would be automatic with the RBD.

[Vaidas Urba]

Nick,
It is the same general theory, but in a different coordinate space. Your 100% gold reserve bank is targeting an overnight cash lending rate that is equal to gold lending rate. This is a clumsy way of describing it, but it is mathematically equivalent.

Of course, at ZLB the quantity coordinate space has an advantage - your reaction function may have a simpler form, as you have a simple form of forward guidance built in. In an interetst rate space you must specify the forward guidance separately. And of course, you get extra credibility for your interest rate reaction function if you do QE.

We have two great natural experiments going on. The ECB is trying to avoid doing QE, it is enhancing the forward guidance every meeting instead to build the credibility. The Fed is trying to divorce the credibility of its reaction function from QE.

[circuit]

Nick, as usual always enjoy your posts. You're a good ambassador for us here in Ottawa.

Re the two theories, Bob Hetzel doesn't see anything wrong in changing his mind about how the Fed is operating. Any chance you could share your thoughts on a piece I did on Hetzel and the money multiplier model? He seems to situate himself somewhere in between the MM and endogenous money folks:

http://fictionalbarking.blogspot.ca/2013/11/on-irrelevance-of-money-multiplier.html

[Nick Rowe]

circuit; thanks!

The money multiplier doesn't play a big role in my thoughts. What I like about it is the disequilibrium hot potato process story. There are two hot potatoes: central bank money circulating between commercial banks; and commercial bank money circulating between people. Plus the distinction between what is true for individual banks vs the banking system as a whole. Plus, most importantly, the idea that money get created when people want to borrow money but do not want to hold money.

Take the orthodox view that the central bank sets a rate of interest. Start in equilibrium, then suppose the central bank cuts the interest rate on borrowing reserves. Individual banks want to expand loans and deposits, because the cost of borrowing reserves has just fallen. And the deposits they create get spent, and hot potato around the people who bank at different banks, and the reserves hot potato around the commercial banks. Just like in the textbook story. The only difference is that the process does not stop until the central bank makes it stop by raising the rate of interest again. And in that new equilibrium, if desired reserves, at the same rate of interest, are some ratio r of deposits, we get exactly the same multiplier (1/r) for the change in M1 divided by the change in MB.

[circuit]

Works for me. I pointed to Hetzel because he basically uses the same language as the endogenous money folks (at least in regard to the operational issues -- even arguing against the relevance of the textbook money multiplier model when tbe CB uses an interest rate operating procedure. Anyway, I have no big issue with what you say. Hetzel, of course, interprets how QE 'works' the same way. I just find it interesting his take on the operational issues is pretty much in line with the post-K (except with respect to the post-2008 Fed; it seems his claim regarding the relevance of the multiplier model is going too far). Interestingly, the BoE paper seems to support Hetzel's view; they claim the "first leg of the multiplier model" is relevant for QE. I'll buy that.

[lxdr1f7]

"But hang on! You have just agreed (sort of) that the demand for base money is some proportion r of the stock of broad money. So in equilibrium, when the actual stock of base money is equal to the quantity of base money demanded, the stock of broad money must be a multiple 1/r of the stock of base money. And if the central bank shifts the supply function of base money $1 to the right, that must increase the equilibrium stock of broad money by $(1/r)"

The S only shifts to the right when broad increases. The impetus is from the banking system. If broad doesnt increase then neither does S.

[Tom Brown]

Nick, thanks. Regarding your statement here:
"It does not matter whether the central bank buys a bond, or a computer, or whatever. Hell, it could just give the money away to its favorite charity (helicopter money), and the result would be the same."

A bit ago I asked Sumner if state sanctioned counterfeiting was equivalent to buying bonds:
http://www.themoneyillusion.com/?p=26213#comment-319935

Scott’s response:
http://www.themoneyillusion.com/?p=26213#comment-319940

"Tom, It’s better to buy the bonds, counterfeiting makes the economy less efficent, as taxes must rise to cover the cost."

Would your "helicopter drop" idea make the economy less efficient too? Would taxes need to rise to cover the cost? If not, why not?

[Max]

If a central bank privatized paper money, and made reserves unnecessary for routine purposes, then perhaps it could shrink its balance sheet to zero. There would be no "money multiplier". Would this upset any economic theories? I don't think so. The money multiplier concept isn't essential to how the system works, it's just a fact about the system, which by over-emphasis serves to confuse things.

[Tom Brown]

shoot, another one in spam. I got rid of the URL and changed emails back to one that worked... so hopefully that helps.

OK, on to a new comment, you write:

"Here is my general theory: when the central bank buys something, with central bank money, the money supply expands."

Counterexample: The commercial banks have near zero shareholder equity and way more reserves than deposits, and the central bank buys all the banks and merges its operation with theirs, and thus its balance sheet with theirs. The reserves go away (cancelled on the one consolidated balance sheet). The bank deposits might remain (now direct CB deposits), but the money supply just contracted. Oops. I guess they have to be careful about what they buy. :D

[Ralph Musgrave]

Nick says “Hell, it could just give the money away to its favourite charity (helicopter money), and the result would be the same.” Actually there is a difference between buying assets and fiscal stimulus funded by new money. Buying assets is distortionary: it relies on just one section of the population spending more, that is, the asset rich. In contrast, fiscal stimulus can be effected in more or less non-distortionary way: e.g. tax cuts for everyone and/or incresed spending by every government department.

As distinct from buying the assets of asset rich INDIVIDUALS, having the central bank buy dodgy assets of banks which are in trouble (lender of last resort) is also distortionary. That is, if banks can have lender of last resort facilities, why not everyone else?

[Tom Brown]

... the weird thing to me about my counterexample above (which I've resurrected from a previous post) is that though the stock of MOA changed, this has no effect on long term prices. Why? MOA net decreased, but so did MOA demand, thus leaving pressure on long term prices unchanged. That's just the explanation in reverse from what you gave me when the only real difference in the example was that the bank deposits exceeded reserves. Recall? In my previous case reserves were $1 and bank deposits $10 prior to the CB taking over. MOA went from $1 to $10, but you said there'd be no effect on prices because demand would increase "10 fold." My broader question is are there any other OMOs the CB might make which change the stock of MOA, but also the demand for MOA, thus leaving prices unchanged? Or have I come up with the only weird example which does that?

[Tom Brown]

... or is that right? Are the two examples just complements of one another?

[Nick Edmonds]

"See that bit about "in exchange for other assets"? That means the Bank of England buys something. Just like I said in my general theory. The central bank increases the money supply by buying something."

I think this is referring to short-term repo actually. The BoE supplies reserves on a day to day basis by lending them against high quality collateral at bank rate, not by buying assets.

[Frances Coppola]

One of the weaknesses of the Bank of England's paper is that it does not distinguish between the pre-2008 world, where bank lending creates demand for reserves, and the post-2008 (QE) world, where the quantity of reserves is entirely disconnected from bank lending and is far more than the quantity required to facilitate deposit movements across the banking system as a whole. Consequently it can appear as if there are two theories of money in this paper. I think it is more accurate to say that there are two different price mechanisms. The pre-2008 version uses the offer price of reserves to control inflation. The post-2008 version uses the bid price.

In the pre-2008 world, OMOs were used to adjust the quantity of reserves so as to maintain the desired offer price: broad money creation was the province of banks. In the post-2008 world, OMOs (QE) are used (ineffectively in my view) to increase the amount of BROAD money in circulation directly, bypassing bank lending. This destroys the reserve demand price mechanism, and forces banks to place money on deposit at central bank, for which they are paid at a rate that the central bank determines. This is why the bid price (IOR rate) is now the primary price mechanism.

The difference between the pre-2008 and post-2008 worlds makes it appear as if there are two different mechanisms by which broad money is created. But actually the central bank has always been able to influence broad money directly to some extent. As Nick points out, under normal circumstances OMOs do have some (small) effect on broad money, not because they increase bank lending (they don't) but because of the substitution of money for assets. Since 2008 OMOs (QE) have been used as a partial substitute for bank lending as the primary driver of broad money creation. But QE is nowhere near as efficient as bank lending as a driver of broad money creation - which is why central banks have to do so much of it. I think this is at least partly why the Bank of England is no longer doing QE and is using quasi-fiscal measures (FLS, H2B) to persuade banks to lend. The paper doesn't mention this.

[JKH]

Nick,

“Paying interest on reserves is just a way to increase the demand for reserves, rather like required reserves.”

With QE, the demand curve for reserves is infinitely elastic – once the quantity of reserves supplied increases to the point where it moves past the zone of inelasticity for uncompensated reserves.

Moreover, changes in IOR from that point shift the demand curve – but won’t change the shape – it remains infinitely elastic.

That also holds for IOR = 0

So IOR doesn’t change the demand for reserves in that context.

The zone of inelasticity represents the pre-2008 Fed for example.

The infinitely elastic zone represents QE.

[JKH]

circuit,

"Interestingly, the BoE paper seems to support Hetzel's view; they claim the "first leg of the multiplier model" is relevant for QE."

There is no multiplication in the first leg of the multiplier.

The first leg of the multiplier is a basic transaction step that is not unique to the multiplier.

So that's not really a demonstration of the relevance of the multiplier - in fact - its more of an intersecting coincidence that provides no special support for the multiplier concept.

[JKH]

The CB prevents rates from dropping to 0 by:

a)Paying IOR = 0, while restricting the supply of reserves to the zone of demand inelasticity (pre-QE)such that the funds rate remains on target

b)Paying IOR = target, while allowing the supply of reserves to move into the zone of demand elasticity (QE)so that a floor is set for the funds rate (ex minor technical market inefficiencies)

It's all about rates

[Mike Sax]

Hi Nick. A few observations. The BOE's 'special theory' sounds an awful lot like MMT theory-they even speak of 'modern monetary theory.'

I saw Sumner not liking the word 'modern' but what is clearly modern is the fiat money system since the Nxion Surprise. That's the idea behind the MMTers anyway. Perhaps the BOE was thinking something like that. It's amazing that this is Mark Carney's CB talking not Warren Mosler.

[Richard W]

I basically agree with Frances here. In a pure quantity managed system, OMOs both have a small direct effect on broad money but a much larger indirect effect via the *price* of reserves. But once you move to a floor system (IOR/FRFARRF), the two become divorced and you just have the small direct effect.

I haven't changed my mind about how this results in a communication headache for central bankers. In this world, the CB doesn't so much have an interest rate reaction function as lots of different reaction functions contingent on various states of affairs, and when you try to explain what a particular future path of rates means for the economy you usually end up in a right muddle. Lower rates now lead to higher rates later, higher rates now lead to lower rates later, and everyone's confused. It would be so much better to say "more money, more NGDP" and rates will go where they need to go.

But so long as the CB's most powerful lever is the price of reserves, not the direct creation of broad money, I'm not sure what can be done about this. And the circumstances where this is a pressing problem - the ZLB - is also a world where the price of reserves is at a natural floor. Even if you're willing and able to do absolutely epic proportions of asset purchases to increase broad money directly, any central bank can easily undo its impact by increasing short rates through other means.

One could - if one was so inclined - choose to conceptualise this as the velocity of money from marginal asset purchases plummeting when the price of reserves is at the floor.

I suppose the question is whether one *ought* to be so inclined. I'm - currently - unconvinced.

[Nick Rowe]

lxdr: "The S only shifts to the right when broad increases. The impetus is from the banking system. If broad doesnt increase then neither does S."

Are you saying that it is impossible for the central bank to have any effect on the broad money stock?

Max: if the balance sheet of the central bank became identically equal to zero, for all values of everything, then it has disappeared. It is no longer the central bank.

Nick E: "I think this is referring to short-term repo actually. The BoE supplies reserves on a day to day basis by lending them against high quality collateral at bank rate, not by buying assets."

That's what I figured they probably meant. But it makes no difference. Making a loan is just buying an IOU. Loans, repos, buying a bond with no repo, are all examples of buying things. When the bond matures the money supply contracts, just like when the loan is repaid. The bank is now selling.

Frances C: when a bank makes a loan of money, and that money gets spent to (say) buy a car, there is a substitution of money for assets, just like the BoE authors say occurs with QE. In both cases it's a hot potato. If the banks are (say) capital-constrained, so won't expand, the BoE can just bypass the banks. The magnitudes may be smaller, but qualitatively the effect is the same.

JKH: I would think of changes in IOR as shifting the demand curve for reserves vertically up or down, not horizontally right or left.

[JKH]

Nick,

agree

but that vertical shift doesn't change quantity demanded

and it sounded like quantity demanded when you said "just a way to increase the demand for reserves"

(although you're typically careful about this sort of thing)

[Nick Rowe]

JKH: we normally put quantity demanded and quantity supplied on the horizontal axis. And increase in demand can mean either a rightward or upward shift. An increase in supply can mean either a rightward or a downward shift.

[dannyb2b]

"It assumes that interest rates are a measure of the "stance of monetary policy". If interest rates were an adequate measure of the "stance of monetary policy", the Bank of England would not need QE."

Idont think they implied rates indicate the stance of policy. I think they meant reducing the rate is more stimulative. At any given growth and inflation rate a reduction in rates in more stimulative.

[Nick Rowe]

danny: if you think of monetary policy as setting interest rates, I agree that lowering rates would be stimulative. But if you don't think of monetary policy that way, and think of rates as a consequence of monetary policy, it is not obvious whether a looser monetary policy would raise or lower rates. That is one of the problems with thinking of monetary policy as setting rates.

[JKH]

Nick,

I understand that in general for sloped curves, and maybe for generalized terminology in total, for consistency I guess.

But in exactly what way does a vertical upward shift of a horizontal demand curve increase demand?

I think the issue with QE has to do with horizontal and vertical curves in this context - a case at the limit.

[Nick Rowe]

JKH: "But in exactly what way does a vertical upward shift of a horizontal demand curve increase demand?"

Suppose the demand curve for Nick Rowe's apples is perfectly elastic (horizontal) at a price of $1 (because anyone can buy identical apples at the supermarket at a price of $1). If I cut the price of my apples to 99 cents, the quantity demanded would increase an infinitely large amount. Now suppose I give away a free pear, worth 50 cents, with every apple I sell, that shifts the demand curve up by 50 cents, so it is horizontal at $1.50. If I keep the price at $1, the quantity demanded increases by an infinitely large amount.

[dannyb2b]

"OK. But if the central bank wanted a temporary increase in the inflation rate, and so a permanent rise in the price level, it would need to shift the supply function of base money, to create a permanent rise in the monetary base, and a permanent rise in broad money,"

But it cant always do this, hence excess reserves. What am I missing?

[Nick Rowe]

danny: that's going beyond the topic of this post. But the short answer is: expectations of future policy. There is a big difference between an increase in the base that is seen as temporary, and an increase in the base that is seen as permanent. But if the stance of monetary policy is seen as setting interest rates, it is hard for the central bank to communicate this difference, except, maybe, by promising to keep interest rates "too low for too long". Unless you believe central banks will forever be powerless to increase the future price level, a commitment to increase the future price level would have the same effect of increasing current expected inflation and thus increasing current actual inflation.

[djb]

"What matters is that the central bank is selling central bank money. It is supplying central bank money."

this must be a typo

should say??

"What matters is that the central bank is NOT selling central bank money. It is supplying central bank money."

[Nick Rowe]

djb: no typo. It does seems strange to talk about "selling money" or "buying money". But when I buy apples for money I am selling money, and when I sell apples for money I am buying money.

[dannyb2b]

"Unless you believe central banks will forever be powerless to increase the future price level, a commitment to increase the future price level would have the same effect of increasing current expected inflation and thus increasing current actual inflation."

The central bank is powerless but if banks dont increase broad money also right? The CB can just increase S of reserves which brings down rates and increases lending until rates go to 0%, then it needs the banks to expand broad in order for price level to rise. Unless the CB goes negative on rates. What about debt to gdp? That is constantly rising. For banks to lend rates will have to go more and more negative in order to compensate for lending to debtors with such poor balance sheets.

Affecting price expectations depends on the capacity of the CB to affect inflation. If the system is blocked then it cant affect inflation.

[Maurice Lechat]

"Quantitative Easing" is just a silly new name for the "Open Market Operations" that first-year textbooks have always said was the way that central banks normally increase the money supply."

Hurrah! For a long time now I've been reading about QE and thinking exactly that thought.

A question...when central bankers put out stuff like this is it (1) because they themselves don't see that bank rate setting and QE are just different approaches to adjusting the quantity of money or (2) because for some reason they want the rest of us to see QE as a new policy gadget invented to deal specifically with post-2008 conditions? It's hard for me to believe (1) because macro textbooks (at least the ones I used) actually talk about bank rate policy and open market operations as alternative ways to change the quantity of money and shift the LM curve. But if it isn't (1) then why (2)?

[Frank Restly]

Nick,

I think there is an element of choice to consider. When a central bank decides through open market operations to buy an asset, do the current owners of said assets have a choice in the matter - can they refuse to sell? I don't think they can, but I could be wrong.

Where as, when a central bank decides to change the interest rate at which it will lend money against collateral, they are not forcing anyone to borrow at that rate. They may in fact make no loans at that rate.

[Nick Rowe]

Maurice: yep. Ever since people started talking about QE, I've been thinking "Is there some difference between QE and OMO that I'm just too ignorant to see?" I think everyone is just too scared to ask!

Good question. I don't know the answer. My guess: it's because central bankers have *not* thought of interest rate setting as changing M. They think of interest rates as affecting investment and saving, and exchange rates and net exports, and M as just a passenger along for the ride. So QE, where they are changing M, really does seem very new and unfamiliar to them.

Frank: if the central bank offers to buy a bond, you don't have to sell. (Except in reverse repos, where you have already promised to sell back when you bought.) Exchange is voluntary.

[JW Mason]

From my point of view, QE and OMO are both interest rate policies. The difference is (1) QE targets longer rates and (2) for some (perhaps ideological) reason, the central bank is unwilling to announce its rate target, so it has to try to achieve it directly by securities purchases, rather than via expectations. This makes QE much less efficient.

From my point of view, in modern economies there is no economically meaningful "M" and nothing that corresponds to the hot potato story. But I do agree with Nick that we should be looking for a general account of monetary policy that embraces both QE and OMO, as well as the other forms that monetary policy has taken historically.

[Frank Restly]

Nick,

"Frank: if the central bank offers to buy a bond, you don't have to sell. (Except in reverse repos, where you have already promised to sell back when you bought.) Exchange is voluntary."

I don't think so. If fiscal authority decides to buy back outstanding debt, it can do so on its own or it can deposit revenues with central bank and instruct central bank to buy back outstanding debt. Ultimately, the supply of bonds is dictated by the fiscal (not monetary) authority.

I realize you were speaking to the general case of the central bank buying any asset, but under current arrangements you could be forced into selling your government bond holdings if the government decides to buy them back from you.

[Tom Brown]

Nick, you write:

"The central bank disappears (its balance sheet has $0 on the liability side), and the one commercial bank is now the new (privately-owned) central bank."

What implications does that have for the economy? The new private "CB" is now an organization dedicated to maximizing profits for its shareholders... Hmmm.

And yet there's still this old CB in the background that can buy assets: say they buy them from Person X. They send Person X a check and he deposits it ... the bank suddenly has a demand for reserves to balance their balance sheet don't they? Say by law they have to accept CB checks. And the CB duly credits the bank with reserves as soon as the check is deposited to settle accounts with the bank. The bank should be OK with that, right? It's equity is left unchanged in the transaction, but for that to happen it needs to be credited with reserves.

Do we really have two CBs? One with a public mandate and one trying to maximize profits?

Nick thanks for opening my mind to the world of hypotheticals... they are fun! Lol :D

[Tom Brown]

Nick, last bit on my hypothetical: run it in reverse: start with the CB having $X > $0 in assets (and thus the bank having $X in reserves). Now if the CB sells assets, such that the new dollar amount of assets it has is $(X*epsilon), with epsilon > 0, then reserves go to $(X*epsilon). In theory prices should eventually go to P*epsilon if they start off at P, right? But that implies the CB can make long term prices as arbitrarily close to zero as it likes by adjusting epsilon appropriately. Does that make sense?

(Notice how I cleverly avoided reserves going all the way to zero, and thus the CB losing its status as the sole central bank)

[soma]

""Quantitative Easing" is just a silly new name for the "Open Market Operations""

traditional OMO: temporary repurchase agreements, counterparty = commercial banks (primary dealers)

QE: permanent purchase, counterparty = commercial banks (primary dealers) + secondary market participators

The FED broadened their definition of OMO and included QE as another liquidity providing operation/tool to fulfill monetary policy objectives.

[chris herbert]

Banks create money by making loans. I'm pretty sure reserves and deposits are next to irrelevant in the decision to make, or not make, a loan. If the Central Bank buys an asset, it isn't creating money, it's just changing the composition. If it just sent money to the banks, and did not buy anything from them, then it would be creating money. At least that's the way I see it. I think this is why people confuse Treasuries with 'debt,' when functionally they really aren't. Paying off Treasuries, or not rolling all of them over when they reach maturity, reduces the amount of financial assets by the amount of 'bank money' that is used to retire the Treasuries. Which is why the US should be very cautious, even timid, in reducing the amount of Treasuries in the system.

[W. Peden]

Soma,

It is conventional in English to only capitalise acronyms and initialisms. 'Fed' is an abbreviation, not an acronym or initialism.

Nick,

Great post! The Bank of England can be very silly at times. Also, the way that one hears talk of "changing interest rates" and "OMOs" as two different ways of doing monetary policy (comparable with, say, altering reserve requirements) is very odd.

[Tom Brown]

chris herbert,

"I'm pretty sure reserves and deposits are next to irrelevant in the decision to make, or not make, a loan."

Not necessarily: simple example: the CB chooses to target a fixed quantity of base money (which is possible for it to do). A bank might be very nervous about extending credit under such circumstances, if reserves were hard to come by. Especially since it has no idea what the interest rate on reserves might be if it had to borrow them. Deposits matter too, in that they are potentially a cheaper funding option (cheaper than borrowing reserves).

"If the Central Bank buys an asset, it isn't creating money"

Sure it is. Reserves are a component of base money (MB). If a non-bank seller sold the asset, then a bank deposit will be credited as well (M1), or that deposit will be withdrawn as cash (MB again). What it's not creating is equity: not for itself, the bank, or the non-bank seller (should there be one).

Treasuries are a liability to the Tsy and an asset to whomever holds them. They represent a debt to the Tsy. Literally, if you hold a T-bond, the Tsy is your debtor and you are the Tsy's creditor. So to the bearer they shouldn't be confused with debt, but to the Tsy they always are debt.

[Mark A. Sadowski]

Excellent post.

Reading through the comments a couple of (your) remarks stand out.

Nick:
"Making a loan is just buying an IOU. Loans, repos, buying a bond with no repo, are all examples of buying things."

This is useful for pointing out that the different operational realities of repos and OMOs are a distinction without any real difference.

Nick:
"when a bank makes a loan of money, and that money gets spent to (say) buy a car, there is a substitution of money for assets, just like the BoE authors say occurs with QE. In both cases it's a hot potato. If the banks are (say) capital-constrained, so won't expand, the BoE can just bypass the banks. The magnitudes may be smaller, but qualitatively the effect is the same."

I liked this point because it demythologizes the uniqueness of commercial bank loans in the broad money creation process. I might add that since the real of cost of base money creation is nihl, the magnitude of the effect is of no real significance whatsoever.

Several of points of my own...

1) "Money Creation in the Modern Economy"

There's nothing particularly "modern" about the current system of money creation. In particular central banks have been doing QE for over 340 years.

2) Banks don't lend out reserves.

This is about as useless a mantra as saying "consumers don't deposit currency", and equally true.

3) Loans create deposits.

When Krugman was asked if loans create deposits, or deposits create loans, his response was "yes".

As he noted, it's a simultaneous system, and in fact when short term interest rates are at the zero lower bound, and central banks are creating base money in ad hoc amounts as with QE, empirically the process is almost always one way from deposits to loans. In particular, Granger causality tests show that it is this way in the US from 1933-41, the US from 2008 to present and the UK from 2009 to present.

4) The money multiplier is dead.

Every Econ 102 textbook I’ve ever seen teaches the money multiplier is a function of three variables. The currency ratio is always portrayed as the depositors’ choice, the reserve ratio (above required, if any) is always the lenders’ choice, and the total amount of currency and reserves (the monetary base) is the central bank’s choice (even if supplied through the discount window), and all are dependent on the conduct of monetary policy by the central bank.

Every textbook that presents the simple model of multiple deposit creation follows this with a critique clearly stating its “serious deficiencies”. In Mishkin’s intermediate level textbook (I have the 7th edition), not only is there such a section, the chapter in which it is taught is followed by a whole other chapter that makes it abundantly clear that the currency and reserve ratios are variables.

All models are wrong, some are useful. The simple model of multiple deposit creation is useful. And there isn’t a single textbook I have seen that doesn’t point out its serious deficiencies. If students pass a course unaware of the simple model of multiple deposit creation’s serious deficiencies, that is the fault of the instructor, not the textbook.

So, in short, there is no such thing as “exogenous money” theory. (Where's the Wikipedia page?) The believers in endogenous money have carefully constructed an exogenous money strawman, complete with a series of erroneous beliefs and nonexistent defective textbooks, so that they could have something to verbally abuse and physically beat up in socially binding demonstrations of rage.

In the final analysis, if the money multiplier doesn't exist, then why does so much Post Keynesian empirical research on the endogeneity of money use the money multiplier as a key variable? (e.g. Thomas Palley, Robert Pollin etc.) Saying the money multiplier is dead makes about as much sense as saying the sectoral sectoral balances is dead. It's an accounting identity for heaven's sake.

5) Nick: "I'm sure they are not alone in having two theories: one for "normal times"; and one for QE, which is seen as needing a special theory only applicable in "abnormal times". But it is rather peculiar having two different theories of the same thing."

What's really peculiar is that the Fed, the BOJ and the BOE have all been in "abnormal times" now for over half a decade. How long before abnormal times becomes normal and normal times become abnormal? (Just wondering.)

Maybe what we need is one general theory of money creation to rule them all.

6) This post vaguely reminds me of another post you did five years ago:

http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/03/the-return-of-monetarism.html

"As interest rates approach zero, and central banks look at "unorthodox" monetary policies, the Neo-Wicksellian perspective on monetary policy has switched to a blank screen. We are witnessing the return of Monetarism.

That's the main reason why economists find it hard to think about unorthodox monetary policies. The dominant Neo-Wicksellian paradigm which fills our heads can say nothing about them. We are forced to return to older, half-forgotten ways of thinking. There is perhaps a "Dark Age" in thinking about monetary policy, just as much as in thinking about fiscal policy.

The dominant paradigm for monetary policy over the last decade has been the Neo-Wicksellian perspective: monetary policy is the central bank setting a short-term nominal rate of interest..."

"...So we need to revert to an older, earlier way of thinking. Monetary policy is about changing the stock of money. The objective of monetary policy, in a recession, is to create an excess supply of money. People accept money in exchange for whatever they sell to the central bank, because money by definition is a medium of exchange. But they don't want to hold all that money. Or rather, the objective of the central bank is to buy so much stuff that people don't want to hold the money they temporarily accept in exchange. An excess supply of money is a hot potato, passing from hand to hand. It does not disappear when it is spent. It spills over into other markets, creating an excess demand for goods and assets in those other markets, increasing quantities and prices in those other markets. And it goes on increasing quantities and prices until quantities and prices increase enough that people do want to hold the extra stock of money.

That's classic Monetarism..."

[Max]

Nick: "if the balance sheet of the central bank became identically equal to zero, for all values of everything, then it has disappeared. It is no longer the central bank."

It's no longer a viable business - it can't make a profit. But it can still exist as a bank with no customers (with government subsidy) and can still successfully target 2% inflation or whatever. It doesn't need a money multiplier.

[Tom Brown]

The Market Fiscalist, let me take a crack at that one: If Tsy deficit spends it doesn't necessarily add to either base money or bank deposits. Assume a cashless system for simplicity. Say Tsy sells a $1 bond and Person X buys it. Then Tsy spends the $1 on Person Y. If you trace that all out aggregate bank deposits didn't change in the end, nor did the quantity of base money. Both changed temporarily, but the net effect is a wash. What did change was that the private sector received $1 more in equity and the Tsy lost a $1 in equity.

[jt]

Nick, I really like this post. Most of all because I didn't have to decode a parable of evil wheat drinking aliens which have two theories of motion - x(t) and v(t).

Let's generalize even more: central banks essentially use their balance sheet. Governments essentially use their balance sheet and social constructs (laws). So how is monetary policy different from government?

[Dan Kervick]

Neither quantitative policy nor interest setting policy "works" in any powerful sense. These are both idle forms of macroeconomic scholasticism preached by the High Church central bank faithful. We have spent four years doing little but debating the miniscule policy contours of decline and stagnation.

My sense is that we are close to the time when the public finally gets it, cuts the Gordian knot and puts an end to these sterile reflections with serious action. An assertive, progressive dirigisme is coming as the many realize that we are once again a developing country with unconscionably barbaric social and economic institutions, and that we can address our economic problems in an entirely direct way rather than looking to baroque and futile finance system workarounds. If our societies need to produce something that the private sector is not producing, the public will produce it; if people need jobs that the private sector is not generating, the public will hire them; if we need to organize the monetary financing of major national projects in education, energy, environment and broad prosperity, the public will use its inherent monetary authority to finance them; if some are struggling and scraping while plutocrats live high and buy politicians, the public will reach right into the bank accounts of the latter to throw them several rungs down the economic ladder, with good riddance.

[dannyb2b]

"If the banks are (say) capital-constrained, so won't expand, the BoE can just bypass the banks. The magnitudes may be smaller, but qualitatively the effect is the same."

What do you mean the magnitudes? The amount of money created? If by qualitative you mean how effective this policy is then I disagree. If money is expanded to corporates that wont spend or just reinvest in other assets this wont have the same effect on demand as transferring money to agents with a higher MPC.

[Nick Rowe]

soma: "traditional OMO: temporary repurchase agreements, counterparty = commercial banks (primary dealers)"

To my ears, that sounds like a strange new definition of "OMO". I thought those were called "repos and reverse-repos with the commercial banks".

Life would be so much easier if we just talked about central banks buying stuff and selling stuff, and then let the Bank's minions deal with the details of what particular stuff they want to buy and sell, like whether they buy a Dodge or Cadillac for the Governor's official car (they bought a Dodge, for optics, and because they liked the name).

chris: "Banks create money by making loans. I'm pretty sure reserves and deposits are next to irrelevant in the decision to make, or not make, a loan."

Nope. Take an extreme example: suppose a bank makes a loan financed by issuing new shares. Unless you call those bank shares "money", no money has been created.

W Peden: Thanks! Yes, but the BoE is certainly not alone in this. When the Bank of Canada talked about QE, it suddenly switched to a whole new (old) way of talking about monetary policy. See my old post here.

Mark: Thanks! And an excellent comment. That old post of mine was the one I wrote when I first discovered Scott Sumner. Reading Scott re-opened my monetarist floodgates, and it all came out in that post.

Max: I don't get it. If a central bank never has a balance sheet (except for the building it occupies), for any value of anything, and so never does anything, and is expected never to do anything, how can it target anything?

jt: thanks!

"Let's generalize even more: central banks essentially use their balance sheet. Governments essentially use their balance sheet and social constructs (laws). So how is monetary policy different from government?"

Central bank liabilities are used as the medium of exchange and medium of account, and all other media of exchange are convertible into those central bank liabilities at fixed exchange rates where the issuer is responsible for ensuring convertibility.

Dan: you are in danger of sounding like the Campus Marxist! Plus, you are ignoring a very long history of central banks and the effects of their (good or bad) policies. Again: what an amazing fluke that the Bank of Canada said it would target 2% inflation, and the average inflation rate for the next 20 years was almost exactly 2%, and very different from the inflation rate in the decades before that announcement. And that is just one example.

[dannyb2b]

"Dan: you are in danger of sounding like the Campus Marxist! Plus, you are ignoring a very long history of central banks and the effects of their (good or bad) policies. Again: what an amazing fluke that the Bank of Canada said it would target 2% inflation, and the average inflation rate for the next 20 years was almost exactly 2%, and very different from the inflation rate in the decades before that announcement. And that is just one example."

IMO you sound pretty Marxist when you say the CB "controls" inflation. Sounds almost like government price controls. In the long term the CB doesn't control inflation. It influences inflation. The CB was able to target inflation while there was a relatively compliant and predictable banking system creating broad money. When that breaks down the CB has a harder time attaining its inflation targets and has to do more extreme things. Eventually no one has control and the whole thing goes south. MP depends on the CB and the private banks in the current system not one or the other alone.

[Nick Rowe]

danny: "IMO you sound pretty Marxist when you say the CB "controls" inflation. Sounds almost like government price controls."

Nope. Price controls are when the government makes a law saying you aren't allowed to buy or sell something above or below a certain price, even if buyer and seller want to. Inflation targeting is when the Bank of Canada adjusts the supply of the good it produces to make the price of the good it produces fall at 2% per year.

[Philip Pilkington]

"But hang on! You have just agreed (sort of) that the demand for base money is some proportion r of the stock of broad money. So in equilibrium, when the actual stock of base money is equal to the quantity of base money demanded, the stock of broad money must be a multiple 1/r of the stock of base money. And if the central bank shifts the supply function of base money $1 to the right, that must increase the equilibrium stock of broad money by $(1/r). Just like the first-year textbook says it will!"

This is wrong. The BoE are recognising that the causality is reversed. The money multiplier assumes that the causality runs 'Reserves=>Lending' whereas the BoE recognises that it runs the other way. This is what endogenous money theorists call the "credit divisor".

http://multiplier-effect.org/?p=9520

On another note, I think that central banks are following us on this one. Obviously you have your own theory and you're sticking to that which is fine. But the CBs aren't going to use your "general theory" they're using ours.

[JKH]

Nick,

"Life would be so much easier if we just talked about central banks buying stuff and selling stuff, and then let the Bank's minions deal with the details of what particular stuff they want to buy and sell"

Sounds like you're including purchase of new production

If so, similar to one version of what I talked about here:

http://monetaryrealism.com/treasury-and-the-central-bank-a-contingent-institutional-approach/

i.e. a consolidation of fiscal and monetary functions and policy into a single central banking institution - I called it the central treasury bank (CTRB)

But I'm not sure you're nearly that open ended about the fiscal side of the minions' responsibilities

(maybe they bought a used Dodge?)

do you have a more limited fiscal effect function in mind as part of it?

I know that the existing CB's budget includes fiscal expenditures and that CB profit is fiscal - but can you flesh out the scope a bit more in terms of your idea here?

[Nick Rowe]

Philip: Nope. It's simultaneous, with demand AND supply, just like in the first year (micro) textbook. The commercial banks and the public create the demand *function* for reserves, and the central bank creates the supply *function* for reserves. And you need both to determine the equilibrium quantity of reserves. And that is true even if one of those functions is flat, in the very short run of the 8 weeks. If the central bank were to shift its supply function up (i.e. increase its target rate of interest) that would reduce the quantity demanded, and so reduce the equilibrium quantity.

"Demand" and "supply" are not points. They are curves, or functions. Again, you really do need to make the distinction, that pedantic teachers of intro economics always insist on, between demand and quantity demanded, and supply and quantity supplied.

The *only* case where you can say that quantity is determined by demand alone is if the demand curve/function is vertical, and the authors are certainly not assuming that, because if it were true the BoE might as well close down. (Similarly, the only case where you can say that quantity is determined by supply alone is if the supply curve/function is vertical, which it would be if the BoE wanted to make it vertical, but it usually doesn't, so it usually isn't.)

JKH: Yep. My little general theory works exactly the same whether the central bank buys newly-produced goods, old goods (can we even tell the difference between new gold and old gold?), or IOUs written on bits of paper. But apart from computers and economists, and fixing up their building, the Bank of Canada buys very little newly-produced goods, so we normally ignore it.

[djb]

Dan Kervick

you are talking about fiscal policy, direct investment into projects that redistribute the money supply

and I don't think even taxing the rich will throw them several rungs anywhere, that's the point

but monetary policy helps too

[Philip Pilkington]

Nick: You're being very evasive here. All the textbooks deal with a CAUSAL relationship. They say that an increase in reserves LEADS TO and increase in the broad money supply through the stimulation of lending. Let me give you some examples.
_______

Richard Froyen 'Macroeconomics': "When banks find themselves with excess reserves after a Federal Reserve open-market purchase of securities, they attempt to convert those reserves into interest-earning assets. THEY EXPAND BANK CREDIT BY MAKING NEW LOANS and purchase securities." (p375 - My Emphasis).

Paul Samuelson & William Nordhaus 'Economics': "We can see that there is a new kind of multiplier operating on reserves. FOR EVERY ADDITIONAL DOLLAR IN RESERVES PROVIDED TO THE BANKING SYSTEM, BANKS EVENTUALLY CREATE $10 OF ADDITIONAL DEPOSITS OR BANK MONEY." (p493 - My Emphasis)
_______

I could provide many more examples but I thought two "Keynesian" textbooks might provide a good perspective on this.

Now, in both of the highlight parts of those quotes we see a CAUSAL argument operating. The assumption is that if the central bank injects $x of new reserves into the system lending will increase by a multiple of this. This is incorrect and the BoE has now officially recognised that it is incorrect.

You can engage in obfuscation all you like and talk about supply-curves and demand-curves but you're not fooling anyone. The textbook money multiplier theory is wrong. The new version of the theory that you put forward has been around in Post-Keynesian circles since the 1980s. It's not the money multiplier, it's called the 'credit divisor'.

Finally, here is Wikipedia on this -- showing that the consensus is on our side on this one.

Wikipedia: "There are two suggested mechanisms for how money creation occurs in a fractional-reserve banking system: either reserves are first injected by the central bank, and then lent on by the commercial banks, or loans are first extended by commercial banks, and then backed by reserves borrowed from the central bank. The "reserves first" model is that taught in mainstream economics textbooks, while the "loans first" model is advanced by endogenous money theorists." [http://en.wikipedia.org/wiki/Money_multiplier]

[Nick Rowe]

Philip: "Nick: You're being very evasive here. All the textbooks deal with a CAUSAL relationship. They say that an increase in reserves LEADS TO and increase in the broad money supply through the stimulation of lending. Let me give you some examples."

Of course they do. And they are right. If the central bank shifts the supply curve, that causes the equilibrium quantity to change (unless the demand curve is vertical, which it isn't). Similarly, if the commercial banks and public shift the demand curve, that too causes the equilibrium quantity to change (unless the supply curve is vertical, which it may or may not be, depending on what the central bank is targeting, and whether the shift in the demand curve would cause the target to be missed).

That is not being "evasive" Philip; that is understanding very very basic first year economics, of supply and demand.

[Philip Pilkington]

Nick, are you therefore saying that if the central bank increases the amount of reserves in the system then the banks will increase lending by an exact multiple of this increase in reserves? Please give a 'yes' or 'no' answer to this question without a foray into supply and demand curves.

[JKH]

Nick,

“If the central bank shifts the supply curve, that causes the equilibrium quantity to change (unless the demand curve is vertical, which it isn't). Similarly, if the commercial banks and public shift the demand curve, that too causes the equilibrium quantity to change (unless the supply curve is vertical, which it may or may not be, depending on what the central bank is targeting, and whether the shift in the demand curve would cause the target to be missed).”

Two entirely separate issues – those curves as they apply to EXCESS reserve BALANCES at the CB, and the multiplier.

As I noted earlier, the demand curve for CB excess reserve balances IS nearly vertical under normal OMO conditions, and then it kinks to horizontal under QE conditions.

The supply curve for CB excess reserve balances IS vertical under normal OMO conditions, because the CB is targeting the intersection of a nearly vertical demand curve and a vertical supply curve – a target intersection at the target fed funds rate. It shifts its vertical supply curve as required to do that.

The supply curve is also vertical under QE – the CB is just targeting the supply of QE for whatever reasons it wants.

None of that has anything to do with the money multiplier. The money multiplier is inoperative – except in reverse where the required ratio is non-zero (i.e. not in Canada) – i.e. deposit creation determines required reserves. And required reserves have nothing to do with excess reserves in the context of the curve analysis for excess reserve balances.

[Nick Rowe]

Philip: No. I would need to change the question, and make additional assumptions, before I would answer "yes".

But instead I'm going to see Tim Horton.

Now, I know you have read a lot of very fancy stuff, but have you ever really understood demand and supply curves? Please give a 'yes' or 'no' answer to this question, without a foray into re-switching.

[Philip Pilkington]

Yes Nick, I do understand supply and demand curves. See? I can answer a question straight even when its rhetorical.

Tell me this: are you going to give your students a chance to evaluate both views on their own? Now that our view is the view endorsed by central banks are you going to give it some space in your classes?

I'm just concerned that you might hobble your students by teaching out-of-date monetarist views when the central banks are making explicit statements that contradict these. Of course you're entitled to defend your views if you please but it seems only fair that your students should hear both sides of the issue, right?

[Mark A. Sadowski]

Philip Pilkington,
"Paul Samuelson & William Nordhaus 'Economics': "We can see that there is a new kind of multiplier operating on reserves. FOR EVERY ADDITIONAL DOLLAR IN RESERVES PROVIDED TO THE BANKING SYSTEM, BANKS EVENTUALLY CREATE $10 OF ADDITIONAL DEPOSITS OR BANK MONEY." (p493 - My Emphasis)"

I have the 16th edition (1998) of Samuelson and Nordhaus and that sentence can be found on page 480 of my copy. On the very following page it states: "THE ACTUAL FINANCIAL SYSTEM IS MORE COMPLICATED THAN OUR SIMPLE EXAMPLE."(My Emphasis.) It then goes on to state that in actuality depositors may choose to hold currency and that banks may choose to hold reserves above the reserve requirement.

Every single textbook that presents the simple model of multiple deposit creation follows it with a disclaimer of this nature.

[W. Peden]

"without a foray into supply and demand curves"

Yes, let's not get too technical here!

[ATR]

Let me try to mediate between the two sides here.

Talking in supply and demand curves for reserves is fine. You can use them to figure out where the interbank rate will land. The central bank can shift the supply curve along a downward sloping demand curve to achieve various different interbank rates.

What Nick might be trying to say is that if the central bank increases the supply of reserves, this should (all else equal) lower the interbank rate, which should then lead to further lending and thus more deposits (simply because the funding rate is lower). I hope both sides can see how this may be a reasonable argument. Where the disagreement may lie is whether the intuition behind the textbook money multiplier has anything to do with how much lending would expand in this case.

But the more monetarist side should acknowledge that the quantity of reserves does not have to increase to lower the interbank rate. The central bank can simply change the corridor by changing its floor (IOR) and ceiling (LLR) rates. It can also change the reserve requirement (although this method isn't so much used in practice).

[Tom Brown]

Nick, you must be sick of my hypothetical. I don't blame you, so I turned to pestering Scott with it:
http://www.themoneyillusion.com/?p=26355&cpage=2#comment-323558

One simple question for you: if there were two or more commercial banks (instead of just one), would the CB go back to being a CB in your opinion (keeping all my other assumptions)?

Thanks!

[ATR]

For those that scoff at those who resist 1st year supply and demand models, I challenge you to step your game up and actually learn about the models that capture the microstructure of these markets. Start with Woodford - http://www.columbia.edu/~mw2230/JHole01.pdf or Bindseil - http://www.macroeconomics.tu-berlin.de/fileadmin/fg124/bindseil/Berlin_lecture_notes_2013_-_hand_out.pdf .

The reserve requirement ratio only provides an anchor around where there is elasticity in the demand curve for reserves. It gives the central bank a target zone where they can alter the supply of reserves to alter the interbank rate. It says nothing directly about how much banks will increase/decrease their lending when their quantity of reserves change.