No it isn't.
History might appear to be just one damn fact after another, but it's the job of social scientists to make sense of those facts, and try to explain the endogenous facts in terms of (relatively speaking) exogenous facts.
Americans might be forgiven for thinking that monetary policy is just one damn interest rate after another, because nobody understands what the Fed is trying to do at any longer horizon. Canadians have less of an excuse. We know what the Bank of Canada is trying to do, because it tells us. Like many other modern central banks, the Bank of Canada targets inflation. The Bank of Canada does not have an interest rate target, except for temporary 6 week periods between Fixed Announcement Dates (and even then, it will change the interest rate target between FADs if it really needs to). It targets 2% CPI inflation. The Bank of Canada's interest rate "target" is an endogenous variable, that can only be understood as determined by the 2% inflation target plus what is happening in the rest of the economy. The Canadian money supply can also only be understood as determined by the 2% inflation target plus what is happening in the rest of the economy.
"The stock of money is demand-determined at the interest rate targeted by the central bank."
No it isn't. But I can understand someone saying that if they also thought that monetary policy is just one damn interest rate after another. It doesn't make sense to say that if you think of the interest rate as an endogenous variable that follows from an exogenous inflation target. (Actually, I think it wouldn't make sense to say that even if the interest rate were exogenous, but then I'm a very heterodox disequilibrium monetarist who has weird beliefs about hot potato money and the law of reflux being wrong, and I'm just going to pretend I'm an orthodox New Keynesian for this post).
Whether a model is useful or not depends on what question you are asking. It depends on your perspective.
Zoom in for a close-up. Suppose you aren't interested in anything that lasts more than 6 weeks. Assuming the central bank has an exogenous interest rate target is then a good approximation to reality. The Bank of Canada really doesn't like to change the overnight rate target unless something big happens, and the actual overnight rate stays fairly close to target, so the approximation isn't bad, as long as we are talking about small changes.
Suppose the demand for bank loans increases. It doesn't matter why. And commercial banks satisfy that demand. The stock of money in circulation increases as a result. It doesn't matter if the quantity of reserves and currency demanded increases too, because the central bank will allow those stocks to increase by an amount equal to the quantity demanded. Does this mean the stock of money is demand-determined? Absolutely not, because the demand for bank loans is a flow demand for loans, and the demand for money is a stock demand for the medium of exchange, and those ain't the same thing, and just because someone wants to accept money in exchange for an IOU doesn't mean he wants to hold a bigger stock of money, it means he probably wants to spend it instead, so there's a disequilibrium hot potato process in which desired expenditure of money exceeds expected receipts of money and…..Ooops! Sorry! I said I was going to pretend I'm an orthodox economist, so yes, the stock of money is, ahem, demand-determined.)
Now zoom out. Let's take a longer, wider perspective.
Suppose the demand for bank loans increases. It does now matter why. Suppose it's due to increased demand for investment. With a fixed inflation target the rate of interest will have to increase, because if inflation was on target before it won't stay on target if desired investment increases and increases aggregate demand too. What happens to the stock of money? It will probably decrease. Take the standard assumption that the demand for money depends positively on the price level and real income, and negatively on interest rates. Then the rise in interest rates means a lower stock of money demanded, and hence a lower stock of money.
Now let's zoom right in very very close. Lets get right down to machine language, and ignore the higher-level programming languages. What do central banks really really control? They control their own balance sheets. And they can communicate promises about how they will change their balance sheets in future conditional on what happens. (OK, some central banks can control required reserve ratios, but I'm going to ignore them here.) If you really want to get down to the nitty gritty, that's it. Central banks control their own current and promised future balance sheets. Any influence they have on interest rates, and inflation, and anything else, all ultimately derives from changing the size and composition of their own balance sheets. Targeting interest rates is not what central banks really really do, for you people of the concrete steppes. Interest rate targets are merely a communications device (and a bad one at that), that some central banks sometimes use (see Paul Krugman) as an intermediate step between the machine language of balance sheet quantities and the ultimate target of inflation (or whatever). Nobody should confuse a communications device for ultimate reality.
Worthwhile Canadian Initiative 
Nick, from my perspective, you are not taking the argument through to the logical conclusion while also changing the goal posts of the debate, in order to defend your viewpoint.
wh10 | April 2, 2012 at 8:02 am | Reply
To preempt, because I believe I have been down a similar road with Nick Rowe before, my sense is that he is saying the interest rate becomes endogenous because the CB will need to react to endogenous market forces in a very specific way to maintain price stability / full employment / or whatever the CB’s goal are etc. In other words, it will be ‘forced’ to alter the interest rate to specific levels to achieve those goals. In my mind, though, even if we assume Nick is right about the influence and necessary path of the CB’s policy, the interest rate is still exogenous because the CB can still adjust it to wherever it likes, regardless of the impact on the economy. But now we’ve entered into a debate about the influence of CB policy and what constitutes good CB policy and have exited a discussion about the ‘exogeneity’ of the interest rate on reserves.
Would you agree?
Scott Fullwiler | April 2, 2012 at 10:26 am | Reply
Yes. There’s nothing that I’ve said anywhere that precludes the CB from running a Taylor’s Rule-type strategy. This is about tactics, not strategy. Tactically, the interest rate is the control variable, not reserve balances, etc. Strategically, the CB sets the target where it wants, responding to events as in a Taylor’s Rule if it chooses. Nick would say it has no choice but to do that–I would just say nothing I’ve said says it can’t do that.
And Fullwiler’s update:
“Update: Paul Krugman has posted a reply to this post that is a straw man. He and Nick Rowe are viewing this all through the lens of the old Monetarist/Keynesian debates in which there was a choice b/n interest rate targets and monetary aggregate targets; the Monetarist critique assumed the Keynesians were going to keep interest rates at the same level forever and not change them. Once John Taylor came up with his “rule,” everyone agreed an interest rate target could work.
What we are talking about here is operational tactics–the CB can only target an interest rate. It cannot target a reserve balances or the monetary base directly. But that is different from strategy–that is, WHERE the CB puts its target and WHEN it chooses to change the target. There is NOTHING in anything I’ve ever said or anything any PK’er, MMT’er, etc., has ever said that suggests the CB can’t set the target wherever it wants whenever it wants. The point is that whatever the target is, THAT is what its daily operations defend directly, not a monetary aggregate, not the monetary base, not reserve balances. There is nothing in anything I’ve said that would preclude the CB from running a Taylor’s Rule type strategy, for instance, that responds at any point in time endogenously to the state of the economy. That is, the target rate is an exogenous control variable (i.e., it is necessarily set by the CB) that it sets endogenously in response to economic events.”
“Targeting interest rates is not what central banks really really do, for you people of the concrete steppes. Interest rate targets are merely a communications device (and a bad one at that), that some central banks sometimes use (see Paul Krugman) as an intermediate step between the machine language of balance sheet quantities and the ultimate target of inflation (or whatever). Nobody should confuse a communications device for ultimate reality.”
Also this high falutin language you are using to continue avoiding thinking about what CB tactics REALLY REALLY ARE in 6 weeks or over the long term. You are simply asserting the qty size of the balance sheet as being the fundamental causal mechanism. Fullwiler has given extensive explanations why its about price, always, regardless of the term considered. I have posted numerous references to this point in the other thread, with regards to how banks change price of reserves – addressing why the intuition from your Chuck Norris analogy is wrong. All I see you offering are assertions and thought experiments that miss the point, supplemented with a “really really.”
Sorry- thanks for the debate. It was stimulating. But I think both parties have hit a brick wall, and I wanted to express my final thoughts.
Nick: Lets say the natural rate rises, but the policy rate is unchanged. So demand for loans increases. And maybe the quantity of money rises, but maybe it doesn’t! Banks might fund the extra loans entirely in the capital markets. The bank of Canada has already eliminated reserves; they can eliminate paper money too if parliament agrees. Nothing whatsoever will change on the day that happens. We don’t need a quantity of the medium of exchange for there to be a price level. We just need a unit of account and any quantity of price stickiness no matter how small. We should just get on with eliminating currency so we can get this argument over with.
K: what would the Bank of Canada’s balance sheet look like in your scenario? Would it be all zeros?
I don’t know about the BoC, but the Fed pays interest on reserves, and my understanding is that it intends to keep the IOR rate close to the federal funds rate target. So if you zoom in very close, the interest rate is more than a communication device. The Fed does have potential control over its own balance sheet, but it chooses not to exercise that control: it chooses to be passive and let private banks expand and contract its balance sheet according to their preferences. Is it true that “…any influence they have on interest rates, and inflation, and anything else, all ultimately derives from changing the size and composition of their own balance sheets”? Not if “changing” means actively changing.
For the zoom out view, I think I agree with Nick, though, but I’m not sure how much I agree with the real Nick vs. the for-the-sake-of-argument New Keynesian Nick.
Nick, central bank controls short-term (O/N) rate. Interest rates on loans, esp investment loans, are long-term rates. If central bank fights inflation today by raising interest rates, long-end of the curve may go down on such expectations causing a further increase of demand for loans.
Andy: Interest on reserves is a promise to adjust the size of a commercial bank’s credits on the central bank’s balance sheet by r% per year until the next FAD (or until further notice). It’s a promise about the relation between current and future balance sheet entry sizes.
“There is NOTHING in anything I’ve ever said or anything any PK’er, MMT’er, etc., has ever said that suggests the CB can’t set the target wherever it wants whenever it wants.”
Well, when a PKer says there is a liquidity trap, does that not suggest the CB can’t lower its target?
Nick
For the purpose of what follows, I will speak in absolutes but the nuanced version should be probabilistic and an answer to ‘what is the size of this effect’.
A central bank has control over its balance sheet given an ‘exogenous’ response function if and only if it has control over the economy/ credit creation. I always understood the PK/MMT position as saying that the central bank does not have much control over credit creation. This was illustrated by assuming an exogenous response function grounded in institutional reality (fed funds rate/ inflation rate) and then showing the indeterminate nature of the central bank’s balance sheet if it tries to stick to its response function, or by showing how the response function was violated when it tried to control its balance sheet.
In Scott Fulwiller’s latest update to his argument, the claim has been reduced to what a central bank’s OMO desk does. With a Taylor rule targeting Fed whose ability to use the Taylor rule to control credit/the economy is not called into question, we’re back in the New Keynesian Nick’s world.
Nick,
It depends on the severity of the “punishment” for deviating from target in the interbank market. But assuming that reserves earn the same as right now (i.e. a bit less that the policy rate) then about the same as the currently outstanding quantity of reserves (i.e. less than $25M or so) plus the amount of deposits held by the treasury ($2 billionish if I remember correctly). So to first order, zero. So zero economically relevant function of the nominal rate. Only the real rate matters.
“Interest on reserves is a promise to adjust the size of a commercial bank’s credits on the central bank’s balance sheet by r% per year until the next FAD”
Not really. If the rate is less than the policy rate, it’s just a way of not punishing a bank too severely for failing to lend reserves (and, in the case of Canada, for not making the net reserve balance zero). If the reserve rate is equal to the policy rate then I’d agree with Andy that that is a way of saying they don’t care about the size of their balance sheet. Also, banks wont then care if they lend to the CB or another bank (so long as collateral is good). It’s the only way you can do QE, but it’s really bad policy as it causes breakdown of the collateral/repo markets if you can just lend to the CB. Anyways, those markets broke 4 years ago.
Off course you would not want the interest rate to be exogeneous in some GE model of the economy. With respect to reality, what could possibly be exogenous (or is exogenous/endogenous even meaningful concepts with respect to debates about free will, determinism etc).
However, if you want to do a thought experiment about the CB doing something different than in the GE model (i.e. making something exogenous by assumption), and also want your analysis to be somewhat relevant for the real world, you probably should let the CB set another interest rate (rule/path) while leaving the money stock endogenous. One obvious advantage of doing it this way is that you will do the same type of analysis that every CB is doing (or, well, they probably do not even include the money stock in the model – but your analysis would at least be more similar), and you have some empirics concerning what effect these kind of actions can have.
If you instead put the money supply as exogenous, you probably have to assume a completely different institutional framework for it to even make sense.
The power of central banks ultimately derives from the ability to create base money. But it doesn’t automatically follow that creating base money must be how central banks “really” steer the economy. What gives CBs their power and how they operate are separate issues.
I’ve been reading all this back and forth between Keen and Krugman, as well as Nick’s related commentary, as well as the related posts by Frederik, Fullwiler, Mason, Ramanan etc. and all of the many fine comments. And to certain degree my head is spinning but I don’t think anyone has succeeded in changing my convictions. However I do think I learned a little something about Post-Keynesianism.
My general sense of things however is that the Post-Keynesians are acting as though they just discovered the Sun is revolving around the Earth (yes I know Keen has also referred to Ptolomy) and celebrating with a bunch of gotchyas. That’s not to say Krugman hasn’t been a little rude himself but he’s never claimed to be particularily polite.
I managed to read Keen’s paper and have a few observations, and I find myself in partial agreement with some previous observations.
1) Keen’s argument that money is endogenously created by the banking system is based purely on Schumpeter and Minsky. He quotes the scripture like the faithful (which is subject to interpretation) and then moves on. To me this is just pure monetary mysticism.
2) His description of “NeoClassical” (who’s he calling NeoClassical?) interpretations of Minsky is a little thin. It seems to me like he is setting up a strawman argument.
3) His statement that “Aggregate Demand equals income plus debt creation” is jaw-droppingly idiosyncratic. This statement, and others like it, render much of what he writes virtually incomprehensible to all the other major schools of economics (be they Neoclassical, NeoKeynesian, Monetarist etc.) He seems intent on inventing his own school (Keensian Economics?), but my advice is that if he’s going to invent new concepts he shouldn’t borrow old terms that are already well defined.
4) I find his Walras-Schumpeter-Minsky’s Law intuitively interesting but poorly expressed and insufficiently justified. To repeat my previous concern, he needs to define new variables, not implicitly redefine old ones.
5) His blog post entitled “Ptolemy and Walras—Brothers in Arcs” reveals he’s seemingly unfamiliar with Neo-Keynesian Economics. That’s somewhat ironic since I sense that a lot of the Post-Keynesians seem to be offended that “NeoClassicists” are unfamiliar with their literature.
In short if Steve Keen wants be taken seriously by the more orthodox economic community he needs to start using their language instead of missappropriating it. Moreover it would be nice if he displayed a better understanding of the ideas held by the people he’s arguing against.
And a final note. The endogenous money people seem oblivious to the fact that central banks don’t always target interest rates. They can target any damned thing they want to (reserves, MB, MS, exchange rates, inflation, unemployment, price level, NGDP, the price of tea in China, etc. etc.). They’re the CB.
Mark A S,
I don’t think “Post-Keynesians are acting as though they just discovered the Sun is revolving around the Earth”. Post Keynesians have been saying the same thing from the days of Nicholas Kaldor and Joan Robinson. But they have been sidelined. I think Kaldor had a lot of weight in the UK and even advised the government regularly and may have had a lot of weight on the budget.
I think you may be probably somewhat correct here. Post Keynesians have found a lot of fans, given the crisis and the role of the internet, blogosphere etc. So you may see a lot of people realizing – at least behaving as if – they have found something new. From their perspective, they are right because without the crisis and the blogosphere it would have been difficult to come across a school of thought (or schools of thought similar to each other) which appeals to many people.
Also, more communication tools may also have brought heteredox schools and scholars get closer.
There are a few things which may not have gone the Post Keynesians’ way. I think within themselves there are differences in approach, style and varied views on how it all fits together. I think a lot of Post Keynesians wouldn’t agree with Steve Keen for example but are united with him at least for a cause. So you are right – nobody in the PKE will agree with Keen on his definition of aggregate demand 🙂 Nontheless they may agree with a lot of what Keen says.
I really do not feel like I have any skin in this game – which makes it very frustrating to not understand what e.g. Krugman and Nick is objecting to. If I hold any kind of bias, it should be towards believing in/rationalizing anything Krugman says.
Are these three statements true?
1: If we fix the interest rate to some path (and keep the current institutional setting), there is no way of knowing how big the money stock will be in five years.
2: If we fix the supply of money to some path (and let interest rates be determined according to some loanable funds framework), there is no way of knowing what the interest rate will be in five years.
3: We cannot have a fixed interest rate and a fixed money supply at the same time (i.e. in any point in time).
If true, does this not imply that if you set the interest rate, the money supply have to be allowed to vary freely (i.e. be determined endogeneously)?
Or is the point that a certain money supply will cause certain inflation, and whether you choose to control money supply or interest rates in the short term, that will have essentially no effect on transaction technology or velocity of money, so the long term money supply is implicitly controlled through the short term interest rates?
I think I have asked before but if the last thing is the case, what is your definition of “money”.
PS:semantics: CB´s do not “target” the interest rate. If the difference between your aim and outcome is that small, and we are in the realm of social/economic policy, you “control” it.
“Suppose the demand for bank loans increases. It does now matter why. Say it’s due to increased demand for investment. With a fixed inflation target the rate of interest will have to increase, because if inflation was on target before it won’t stay on target if desired investment increases and increases aggregate demand too. What happens to the stock of money? It will probably decrease.”
Dun dun dun! Nick Rowe is making an awful lot of assumptions here. He’s assuming, as far as I can see, an economy operating at full employment (NAIRU?). So, in Rowe’s view any increase across the IS curve will lead the central bank to intervene and shift interest rates. Rowe is just shifting the goalposts here big time.
Dean Baker and I had a fairly sophisticated discussion of this based on the long-term (10 year) interest rate and CBs hiking rates due to inflation expectations. After Greenspan we know that this isn’t very straightforward and there is a LOT of central bank leeway here. They may try their hand at redefining NAIRU or whatever. But even so, Rowe probably shouldn’t take a full employment economy as his case study.
This is all peripheral to the main discussion of course. But my feeling is that this discussion will never stay focused.
Mark A. Sadowski: “Keen’s argument that money is endogenously created by the banking system is based purely on Schumpeter and Minsky.”
Maybe. Didn’t read it. But it’s also standard Woodford. You calling him a mystic? The PostKs and the NKs have essentially no differences as far as the current debate is concerned. It is the MMs and the rest of the monetarists who are “mystics” with their belief in the “money supply.” We can’t define it, we can’t measure it, but gosh darn it, we know it’s there. Beware the velocity!
“The endogenous money people seem oblivious to the fact that central banks don’t always target interest rates.”
Rates (repos) are an instrument, not a target.
The monetarists seem oblivious to the fact that central banks don’t give two hoots about the money supply, never mind target it. They tried once, a long, long time ago, but it didn’t work. They haven’t looked back.
I’d just like to point out from Krugman’s blog:
Nick RoweChelsea, Canada
FLAG
Yep.
Just working on a new post. What the critics are saying is:
“Monetary policy is just one d**** exogenous interest rate after another”.
They need to zoom out a bit.
April 2, 2012 at 12:36 p.m.REPLYRECOMMEND1
SHARE THIS ON FACEBOOK
SHARE THIS ON TWITTER
Scott FullwilerIowa
Wrong, Nick. That’s not what we’re arguing at all. Sorry.
Why put words in the opposition’s mouth, like this?
Also, I think that it should be pointed out that in high inflation scenarios if the central bank raises rates the money supply — or, at least, what certain economists call the ‘money supply’ — continues to rise. If we take this inconvenient fact into account Nick Rowe’s above argument (which is HIGHLY theoretical) falls apart.
It’s quite obvious why this might happen, even if we accept Rowe’s presupposition. If we have an inflation rate of 50% per annum and the central bank fixes interest rates up from 10% to 25%, it is still probably a good idea to take out a loan and spend/invest because you can be sure the loan is going to be eroded by the inflation.
The empirical evidence supports this. Of course, what I just said is a huge oversimplification, but it puts a rather large dent in Rowe’s Canada-centric argument as outlined above.
The great thing about the endogenous theory is that it strips away all the presuppositions and complexities and adds them in later. Rowe and the ISLMic Keynesians start with a model that has lots of presuppositions (NAIRU employment, CB acting in a VERY specific and predictable way etc.) and then tries to pare it down to fit the particular circumstances.
I ask you: which approach looks more like science and which looks more like metaphysics? I don’t think I need to answer that question to anyone who is open to posing it to themselves reasonably…
Ramanan,
Yes, I appreciate the fact that Post-Keynesianism has a long history and several strands. I also appreciate that those who believe that money is endogenously created by commercial banks include not only Post Keynesians but Modern Monetary Theorists and people who are more difficult to classify like Keen.
K,
1) It is most certainly not standard Woodford. Woodford is a Neo-Keynesian and I know no Neo-Keynesian who believes that money is endogenously created by commercial banks. That would imply monetary policy is always ineffective. Some Neo-Keynesians do believe that monetary policy is ineffective in a liquidity trap. Woodford, however, is not even one of those. He has stated unequivocally that it is still possible in principle to achieve substantial stimulus to aggregate demand by changing expectations regarding future monetary policy. In fact he’s written more than one paper on optimal monetary policy in a liquidity trap. Saying Woodford believes that money is endogenous is like saying the Pope is an athiest.
2) I see very little similarity between Neo-Keynesians and Post-Keynesians otherwise there would not be this constant back and forth. Moreover Neo-Keynesians have much more in common with Market Monetarists (or whatever they are called these days) as both do not believe in the endogeneity of money and many Neo-Keynesians, like Woodford, believe that monetary policy is still effective in a liquidity trap.
3) During the period when the FOMC was setting a FFR target that was the target. The target may have been set according to a Taylor Rule but nobody really knows for sure (except for the flies on the wall). And it is precisely because the target was an interest rate that open market operations had to be so active during that period. (IMO that period ended in late 2008). Moreover the FOMC did briefly target money supply but gave it up precisely because velocity is so variable. And there are other central banks besides the Fed. There are a variety of possible targets, not just interest rates.
Attention. Attention.
http://www.nytimes.com/roomfordebate/2012/04/01/how-to-teach-economics-after-the-financial-crisis/economics-101-not-your-fathers-monetary-policy
Alan Blinder on monetary policy (h/t wh10)
Couldn’t be better timed, eh?
@ Sadowski
“It is most certainly not standard Woodford. Woodford is a Neo-Keynesian and I know no Neo-Keynesian who believes that money is endogenously created by commercial banks. That would imply monetary policy is always ineffective.”
I don’t think Post-Keynesians think that monetary policy is always ineffective. The Vaolcker shock was definitely effective enough to cause a recession — and probably for quite real reasons. Most of the time monetary policy seems to work — to the extent that it does work, which may be limited — through expectations. That’s why I think we see a lot of mysticism around central bankers who, frankly, come across as modern day shamen (remember what they used to say about Greenspan?.. Ugh… it was f-ing primitive!).
PKers do think that monetary policy has an effectiveness, but they (we?) don’t think its very effective at all. And certainly to the extent that it is effective it generally works on the ‘animal spirits’ rather than on rational decisions undertaken by investors and businessfolk.
Your point (2) is spot on, I think. I see almost no difference between neo-Keynesians (I call them ISLMists… hehehe) and Market Monetarists.
“Moreover the FOMC did briefly target money supply but gave it up precisely because velocity is so variable.”
I think that this might only be partially true in the case of the US. While in the UK a few monetarists really got the reigns of power, I don’t think it happened in the US. There’s a great interview going around where Volcker says that he always thought money supply targets were voodoo but that he cleverly used the political momentum to push for otherwise unpopular interest rates.
In the UK there was a big experiment and many that partook and were not ‘true believers’ came away bruised. There’s a great interview from the 90s with a central banker who says — figuratively, of course — that he has nightmares to this day that what was really was going on was a conscious effort to expand the ‘reserve army of the unemployed’ to break unions. I remember the first time I saw that my mouth fell open… it was coming out of the mouth of a central banker, but it sounds like something Arthur bloody Scargill might have said.
The investment banks created money (debt) by lowering lending standards resulting in inflating housing prices, but while this added to otherwise deficient demand, it did not create (consumer) inflation for the Fed to fight. When the folly was seen and lending standards were raised it destroyed money (debt) and demand leaving us in a hole and the Fed to struggle against. So I would say money (debt) is endogenous, but not wholly out of control of the Fed who can exert force on both lending standards and money (debt) through rates (debt service) but only if collateralizable (not underwater). Am I even close to what is being discussed?
Mark A S,
Keen’s advocacy of endogenous money is not based merely on the fact that Schumpeter, Minsky, Keynes and whoever else advocated it, but on extensive empirical literature. Read his post here:
http://www.debtdeflation.com/blogs/2009/01/31/therovingcavaliersofcredit/
“With a fixed inflation target the rate of interest will have to increase, because if inflation was on target before it won’t stay on target if desired investment increases and increases aggregate demand too.”
Sure, or people will use the extra money to create a massive housing bubble which is not inflation just a big increase in prices (from BoC perspective).
Ritwick: you lost me a bit there. But I agree that if we take the central bank’s reaction function as exogenous, (for almost any reaction function) then the time-path of the central bank’s balance sheet becomes endogenous.
nemi: “Off course you would not want the interest rate to be exogeneous in some GE model of the economy. With respect to reality, what could possibly be exogenous (or is exogenous/endogenous even meaningful concepts with respect to debates about free will, determinism etc).”
I think that the exogenous/endogenous distinction is as much a statement about models, rather than the variables themselves. For example, (nearly) all economic models treat the weather as exogenous, but all meteorological models treat the weather as endogenous. They can both be right. I think we always have to specify: endogenous/exogenous with respect to what?
Max: “The power of central banks ultimately derives from the ability to create base money. But it doesn’t automatically follow that creating base money must be how central banks “really” steer the economy. What gives CBs their power and how they operate are separate issues.”
I think that’s a good point. But I would emphasis their communication channel, which, at root, is the set of conditional promises/expectations they make about their future balance sheets.
Mark: “I’ve been reading all this back and forth between Keen and Krugman, as well as Nick’s related commentary, as well as the related posts by Frederik, Fullwiler, Mason, Ramanan etc. and all of the many fine comments.”
Who is Frederik? I missed that one. Anyone got a link?
“3) His [Steve Keen’s] statement that “Aggregate Demand equals income plus debt creation” is jaw-droppingly idiosyncratic.”
Yep. But deep down I think he’s maybe trying to get at the same thing I’ve been trying to get at like here. Or I could be wrong.
“And a final note. The endogenous money people seem oblivious to the fact that central banks don’t always target interest rates. They can target any damned thing they want to (reserves, MB, MS, exchange rates, inflation, unemployment, price level, NGDP, the price of tea in China, etc. etc.). They’re the CB.”
Yep. And the one thing we know they can’t ultimately target, ever since Wicksell, is the rate of interest. Which is ironic. (Actually, we also know, ever since Friedman, they can’t ultimately target unemployment either.)
Philip Pilkington: “It’s quite obvious why this might happen, even if we accept Rowe’s presupposition. If we have an inflation rate of 50% per annum and the central bank fixes interest rates up from 10% to 25%, it is still probably a good idea to take out a loan and spend/invest because you can be sure the loan is going to be eroded by the inflation.
The empirical evidence supports this. Of course, what I just said is a huge oversimplification, but it puts a rather large dent in Rowe’s Canada-centric argument as outlined above.”
Yep. I’m well aware of that. It’s called the Howitt/Taylor Principle. (Most people call it the Taylor principle, but Peter Howitt stated it first.) A central bank following an interest rate reaction function needs to raise the nominal interest rate more than one-for-one with (expected) inflation to increase real interest rates and keep inflation from exploding. See my old blog post, and the first comment by David Andolfatto.
Mark Sadowski: “Woodford is a Neo-Keynesian and I know no Neo-Keynesian who believes that money is endogenously created by commercial banks. That would imply monetary policy is always ineffective.”
I really don’t get why you say that. Couldn’t someone believe all these 3 things: commercial banks create money; the amount they create (generally) depends both on what the central bank does and on other things too (so it’s endogenous in that sense); monetary policy is effective? I believe all those 3 things!
Have the fundamental points been resolved yet? Are banks constrained in their lending by the their reserve position? Is it required to include the banking sector in a story about debt and leveraging?
Moving on, Rowe, you’ve made the point repeatedly that the target interest rate that is set by central banks is an endogenous variable. In the post today you argue that it is targeting inflation, so central banks follow some sort of Taylor’s rule, plus another other economic variable (e.g. unemployment) that current theory suggests that monetary policy can influence. My question is, why is this relationship assumed to be a structural one, and not a behavioural react function that is conditional on current theory?
My point of disagreement is that, if you pull the camera back then every action by any agent is ultimately an endogenous variable, everything is connected to everything and determining everything, and any economic agent action by any economic agent that is observing a set of economic variables is also endogenous. The point of arguing that the interest rate is exogenous is because ultimately it is up to the discretion of the central banks what their target will be, their expectations that a change in the target will have or no change in the target will have, upon a set of variables which theory suggests they target. However, even if central banks are placing a heavy weight on inflation, it doesn’t necessarily mean that they will then change the target rate when the rate of inflation moves outside their target band, other factors could be more heavily weighing upon their decision, for instance, a small open economy could be experiencing inflation above their current inflation comfort zone (e.g. 2 -3%) however, factors which are currently occurring overseas (a global recession) could make weigh more heavily on their decision, causing them to not raise rates.
Assuming that interest rates are an endogenous variable, how does this change the Post Keynesian story? In their model banks are still not constrained by the quantity of reserves, but by capital, credit worthy borrowers and profitability of potential borrowers. The central bank continues to accommodate the demand for reserves at its target interest rate, and it is still using a short-term nominal rate as its primary tool conditional on the expected impact that this tool has a set of economic variables?
mdm,
Good points.
Btw …
“The only truly exogenous factor is whatever exists at a moment of time, as a heritage of the past.”
– Nicholas Kaldor, Economics Without Equilibrium, 1985.
UnlearningEcon,
The post you linked to helps explain a lot. I’m still digesting it but let me summarize what I think I’ve learned so far.
First of all read Keen’s Berlin paper. Few of the people defending it seemed to have taken the time. In his explanation of the endogeneity of money relies primarily on theoretical explanations by Schumpeter and Minsky. (Keynes is not referred to at all.) Thanks to your link I noticed he does also refer to a paper by Kydland and Prescott and by extension Basil Moore. Basil Moore is a Post Keynesian who I’m not familiar with. Kydland and Prescott I’m quite familiar with, and to be frank, the fact that his empirical justification for the endogeneity of money depends on that dynamic duo raises the hairs on my neck.
The reason is that they are both Real Business Cycle Economists. Without getting too dismissive, RBCers don’t really believe in the effectiveness of AD management. What’s interesting is this remined me of a paper by Brad DeLong called “The Triumph of Monetarism?” in which he says:
“Twentieth century macroeconomics ends with the community of macroeconomists split across two groups, pursuing two research programs. The New Classical research program walks in the footprints of Joseph Schumpeter’s Business Cycles (1939), holding that the key to the business cycle is the stochastic character of economic growth. It argues that the “cycle” should be analyzed with the same models used to understand the “trend” (Kydland and Prescott, 1982; McCallum, 1989; Campbell, 1994). The competing New Keynesian research program is harder to summarize quickly.”
In short Keens apparently traces his support for the endogeneity of money to the same root from which RBC sprung. And if I had quoted further you would see how much Neo-Keynesianism has with Monetarism. No wonder I disagree with Post-Keynesianism. In some ways it is the opposite of what Keynes believed.
None of this of course addressing the Kydland-Prescott observation that monetary lags do not behave according to what the standard money multiplier model suggests. Without getting into much detail I would argue that monetary policy does not function like some hydraulic mechanism where you press down on a bicycle pump inflating an inner tube. It is much more expectations based and I do not, nor have I ever subscribed to the notion of “long and variable lags”.
Nick Rowe,
1) The Frederik post is here:
The highlighted quotes are supposed to be Post Keynesian gotchyas but if take them in context they aren’t gotchyas at all.
2) With respect to targeting interest I agree, in the long run. All I ever needed to know about pegging interest rates over the long term I learned from this paper:
Click to access 58.1.1-17.pdf
3) Given the other two assumptions, monetary policy is only effective if #2 (“the amount they create (generally) depends both on what the central bank does and on other things too (so it’s endogenous in that sense”) is accepted. But it’s not clear to me that engonenous money crowd believe that assumption.
The Goodhart paper in the Frederik post is a speech called ‘Whatever happened to monetary aggregates’ in which he actually extols the need to look at monetary aggregates.
“Yep. I’m well aware of that. It’s called the Howitt/Taylor Principle. (Most people call it the Taylor principle, but Peter Howitt stated it first.) A central bank following an interest rate reaction function needs to raise the nominal interest rate more than one-for-one with (expected) inflation to increase real interest rates and keep inflation from exploding. See my old blog post, and the first comment by David Andolfatto.”
Here’s my fundamental problem, Nick. And I think its a rather obvious one. Your arguments — how shall I put it — proceed in a scholastic manner. You take a model — a perfect or almost perfect model — and then start adding qualifications that subtract from its perfection. These qualifications have fancy names and win people Nobel Prizes; but 99.99% of the time they are common sense propositions — I pulled my inflation argument out of the air driving home today and checked some simple data to confirm it. (No Nobel for me though… I’ll live!).
The Post-Keynesians are far more modest. They strip away all qualifications and posit a ‘base model’ which to work off. People can then add qualifications as needs be and see how they might impact the ‘base model’. They don’t even need a Road to Damascus moment like I had regarding high inflation (kidding, its not that clever at all… its quite simple really). All they have to do is look at the world around them.
Now, I ask you: which approach looks more like science and which looks like scholasticism or metaphysics? I think the answer to that is obvious. An ideal model with qualifications added gradually to restrict its perfection is metaphysics of the highest order and leads to mass confusion. A stripped down model with qualifications added as they appear in reality is a scientific model.
Make no mistake everyone, that is what is at issue here. Scientists will see the sophistry from the truth. As will methodologists and philosophers. Those already trained will see only the models. Frankly, the difference is between academic elitism and the pursuit of truth. We’ve seen this conflict many times before in history. Enlightenment did away with much of the Doctrines. We’ll see what happens in the coming years. My feeling is that mainstream economics is rotting badly at the moment. But we’ll see. Neither you nor I can steer history.
@Mark A. Sadowski: “In some ways it [post-keynesianism] is the opposite of what Keynes believed. ”
What Keynes have you read and what post-keynesianism have you read?
You’re argument is quite frustrating. Keen uses mainstream empirical literature because he thinks that people will find that easier to digest then heterodox empirical literature. are you actually trying to slam him as an RBC thinker for that? I find it difficult to believe that you see that as arguing in good faith.
Nathan Tankus,
I hold Keynes in high regard. I keep my well worn copy of the General Theory at hand to combat misquoting anti-Keynesians at all times.
Granted, my familiarity with Post-Keynesians is limited. That’s why I’m enjoying this cycle of exchanges. I’m learning (I think).
And I’m sorry if Keens thought quoting Kydland-Prescott would earn him some friends in the more orthodox portion of the economic community. Apparently he isn’t aware their is a very deep schism, and the very mention of RBC makes many of us draw our research papers from our holsters.
P.S. On a less grandiose note, we have now established that Nick’s above statements depend strictly on: (a) an inflation rate lower than nominal interest rate and (b) an economy operating at full output/NAIRU.
What does the endogenous theory presuppose in terms of macrodynamics? Nothing. Zero. Zilch. All it asks is that you can apply it to the very real world around you and understand what effects monetary policy may or may not have etc.
Now, I wonder… which argument would old William of Ockham take his razor to? Hmmm…
And which is closer to science? Again, I will let the reader decide.
On a more serious note, Nick, where is my attribution? 🙂
Back to the topic at hand, we reached the same impasse we’re at now back in this previous post.
I agree with you that the extra money created by bank lending has to go somewhere (reduced velocity or higher prices) and that it could lead to higher prices which would cause the central bank to raise interest rates which cause defaults and a greater flow of interest back to banks and a lower desire to borrow all of which would counteract the initial expansion, but still disagree with your view (implied) that this means we don’t need to be concerned about debt levels, the role played by banks in expanding debt levels, potential instability caused by aggregate debt levels, impacts on asset prices and economic growth and so on. The use of the extra funds created by bank lending to create asset bubbles is just one way bank lending can lead to trouble without an inflation targeting central bank having any impact on the process.
mdm: “Have the fundamental points been resolved yet?”
That had me laughing!
Sorry guys, but I’m feeling a bit burned out, and need to calm down a little and have a quiet night. I’m reading all the comments, but not up to responding to all of them.
Edward Harrison here. I run the blog Credit Writedowns where a number of the Keen and Fulwiller posts ran. I just wanted to commend Nick for maintaining a tone of civility which is exemplary given the snarkiness which pervades many of these monetary policy threads. The Krugman monetary policy threads seem to have been shut down by holding comments in a moderation queue. I hope that isn’t the case because it would be disappointing. Ultimately, laypeople are going to learn a lot more from comment threads like yours, Nick. Kudos.
Edwardnh,
I agree. I had a decision to make today. Where should I post my comments? I found all of the arguments on this topic compelling. I didn’t want to get into a tit for tat. Of all the choices Nick’s blog seemed the best place. The responses to my comments here have been highly informative and very thought provoking.
Still agree with Edward about Nick, as I had noted in the other comment thread.
@Mark A. Sadowski he is aware. does that make their empirical work invalid? I don’t think Keen thinks or thought he would ever win any orthodox friends.
I’m happy to hear that. So do I. Remember though, Keynes wrote books before the general theory and wrote many clarifying articles after. many Post-Keynesians draw on that work also. Endogenous money for example, is generally seen to be not in the general theory (Keynes doesn’t focus on banking there)but more in the treatise on money and (especially) in his post general theory articles and in his correspondence.
I also would like to join the group of people commending Nick for being a facilitator in this argument and being very respectful given the circumstances. i said it in another thread but i thought i would repeat it.
Edwardnh: Thanks. I’m not sure i always deserve your praise though. Sometimes i lose it a little. I feel a bit bad about losing it with steve keen’s post. As someone pointed out, he did sort of qualify what he said a bit further down, though it’s a bit like saying someone’s an axe-murderer, then saying a little later that technically he doesn’t have an axe!
It gets a little frazzling when there are hundreds of comments and most people commenting are saying I’m wrong! And I’m not!!
@Nick Imagine how we feel when we open a textbook 🙂
but seriously, you do deserve it. losing it a little bit a couple times doesn’t count against you in my book.
Philip: “These qualifications have fancy names and win people Nobel Prizes; but 99.99% of the time they are common sense propositions — I pulled my inflation argument out of the air driving home today and checked some simple data to confirm it.”
OK, maybe you’re smart (not meant as snark). A lot of people don’t get that. Joan Robinson was a very smart economist, but she didn’t get that point. She thought that the German hyperinflation could not have been caused by loose monetary policy, because the Reichsbank set interest rates at a “very high” level, like 20% (a couple of million % below the inflation rate)!
You want a simple model, that you can add stuff to later? OK. MV=PY?
Nick, as someone who’s blogged and commented off and on for many years and knows how it can almost inexplicably drive even the politest people to insults and sarcasm and anger, I don’t know how you manage to remain so civil despite being involved in so many contentious blog debates (and people only bother to comment if they think you’re wrong, if you’re right, why bother).
But looking back through your many posts on this topic over the last few years made me appreciate how much your posts and responses to my comments have improved my own clarity of thought on this issue which has always been a source of great puzzlement to me, so thanks.
Nick:”OK, maybe you’re smart (not meant as snark). A lot of people don’t get that. Joan Robinson was a very smart economist, but she didn’t get that point. She thought that the German hyperinflation could not have been caused by loose monetary policy, because the Reichsbank set interest rates at a “very high” level, like 20% (a couple of million % below the inflation rate)!”
I will not argue Joan’s position because i have not read her position. I will however, argue John Maynard Keynes’s position that he made in real time. He (and others) argued that German hyperinflation was a balance of payments crisis. they had an enormous (foreign denominated) debt burden thanks to losing the war (and the u.s squeezing it’s allies for lent money). The debt burden was simply too large to deal with by selling asset. If it tried to export much more, it’s revenue would actually fall(and the allies tended to put up higher tariff barriers in response). If they tried to cut imports too much, exports would fall off more because much of it (including food) was a basic input to production. the money printing was the effect, not the consequence of the hyperinflation.
http://www.jstor.org/stable/2224211
Nick, I’m in agreement with the comments above, thanks for providing a civil place to discuss what is a very heated issue.
Wray has the following blogpost for those interested: http://www.multiplier-effect.org/?p=4218
p.s. hopefully all the warmth and fuzziness of the last few comments will make you feel a little better, Nick! I’m really looking forward to your thoughts on my comment 🙂
It’s alright Nick, I think you’re right … but maybe that’s the problem!
Money is special. If you don’t understand that, then you don’t understand anything. Pretty much all recessions are reducible to a monetary disequilibrium. I even agree that it’s mostly a medium of exchange rather than unit of account issue.
People get confused about money. They know central banks can make inflation, and their models capture that relation, but they can’t explain how it happens. It depends on central banks creating a greater quantity of money than people desire to hold, but that is impossible if money is assumed to be little but a safe, zero-interest bond. For example, a government cannot issue a bond (at a given rate of interest and maturity) unless it can find someone who wants to hold it. If people, retrospectively, find themselves holding more government bonds of some kind than they desire, then the price on those bonds fall or they mature and cease to exist. This is not true for money, because money is the medium of exchange. People will accept money even though they don’t want larger average money balances, and they can only get rid of money by passing it onto someone else. The end result is inflation–nothing else but money has this effect.
If money is conceptualised as just another asset, then it’s impossible for central banks to create a greater quantity of money than people desire to hold. It is assumed that, like a government and its bonds, central banks can only create money if they can find someone willing to hold it, and by that assumption the possibility of monetary disequilibrium disappears. One can assume inflation occurs and even model it, but one cannot explain what is going on.
I’m actually trying to write a guest post for Beckworth about all this stuff.