Do you believe in the money multiplier?

What a mind-bogglingly empty question!  The money multiplier equals the ratio of the money supply (however defined) and the monetary base.  It’s simply a ratio, there’s nothing to believe or disbelieve.  It’s like asking if someone believes in the ratio of men to women.  Or whether MV=PY.

So let’s start over.  Do you believe the money multiplier is stable?   Most economists would answer; “It depends.”  Or how about; “Do you believe the money multiplier is useful?”  Now we are beginning to get somewhere.  For instance, do economists believe that an increase in the monetary base will cause the money supply to rise by an amount equal to the change in the base times the multiplier?  It turns out that the answer is; “No in the short run, but yes in the long run.”

What we actually need to do is start with the concept called “the neutrality of money,” which underlies almost all of macroeconomics, and has done so for hundreds of years.  This says that an increase in the monetary base will not affect any real aggregates, and hence all nominal aggregates will rise in proportion.   This suggests there are “multipliers” for every conceivable nominal aggregate, from nominal spending on toasters, to nominal spending on Brazilian waxes, to NGDP.  The “multiplier” for NGDP has a special name; “base velocity.”  But it actually has nothing to do with the velocity of money (the vast majority of money expenditures are not for final goods and services), and should be called the “NGDP multiplier.”

So according to the neutrality of money, a 10% rise in the base will increase all nominal aggregates by 10% in the long run.  Since M1 and M2 are nominal aggregates, they are affected just like nominal toaster expenditures and nominal Brazilian wax expenditures.  (I keep mentioning the latter in a pathetic attempt to show that I am a hip 56 year old keeping up with new industries that didn’t exist when I was 20 years old.)

Of course wages and prices are sticky in the short run, so a 10% rise in the base does not have a proportional effect on most nominal aggregates in the short run.  The neutrality of money (and the money multiplier) are long run propositions.  However some economists believe the money multiplier is relatively stable when interest rates are positive, even in the short run.  This is because the multiplier has two behavioral components; the reserve ratio and the currency/deposit ratio.  Many economists think that during normal times (when interest rates are positive), banks hold very little in the way of excess reserves.  Hence the reserve ratio is stable, and equal to the required reserve ratio.  They also believe the currency/deposit ratio is relatively stable in the short run.

This leads to the question of whether the money multiplier is useful.  I don’t think it is, but let me try to explain why others disagree.  They think a 10% rise in the base (when interest rates are positive) will result in a roughly 10% rise in M2, even in the short run.  I have doubts, but I’m willing to accept that claim.  Then they claim that changes in M2 have an important causal impact on NGDP—M2 is an important part of the monetary transmission mechanism.  That’s the part I reject.  I think future expected NGDP, plus asset prices, are the key transmission mechanisms.  And future expected NGDP is driven by future expected changes in the supply and demand for base money.  I don’t see anything special about M2.

All economists believe the money multiplier becomes highly unstable at zero rates, mostly because, well . . .  because it does become highly unstable at zero rates.  Yes, sometimes economists actually do believe what they see with their own eyes.  I should add that if IOR is permanent, which seems increasingly likely, the money multiplier may become highly unstable at positive interest rates as well.  Money will still be neutral in the long run, but the usefulness of the money multiplier (which I doubt even without IOR) will become even smaller.  It will be as unimportant as the toaster multiplier, or the . . . Brazilian wax industry multiplier.

PS.  No time to comment on his new post, however I never get depressed reading Steve Waldman.  But I do strongly disagree with this:

I’m intrigued, but I’m kind of an idiot. The rest of you are very nice and smart and reasonable.

Steve’s so nice and smart and reasonable that reading his posts makes me realize that I need to try harder.


Tags:

 
 
 

47 Responses to “Do you believe in the money multiplier?”

  1. Gravatar of Nathan Tankus Nathan Tankus
    8. April 2012 at 18:42

    It appears as if you missed the whole debate. The argument is about the causality between variables in the money multiplier.

    “What we actually need to do is start with the concept called “the neutrality of money,” which underlies almost all of macroeconomics, and has done so for hundreds of years”

    I wish economists with little to no training in the history of economic thought would be a little more humble in making claims such as this.

  2. Gravatar of Major_Freedom Major_Freedom
    8. April 2012 at 18:42

    So let’s start over. Do you believe the money multiplier is stable? Most economists would answer; “It depends.” Or how about; “Do you believe the money multiplier is useful?” Now we are beginning to get somewhere. For instance, do economists believe that an increase in the monetary base will cause the money supply to rise by an amount equal to the change in the base times the multiplier? It turns out that the answer is; “No in the short run, but yes in the long run.”

    The multiplier is not constant. Since there is always a money supply and always a monetary base, there is always a multiplier. So do economists believe that an increase in the monetary base will cause the money supply to rise by an amount equal to the change in the base times the multiplier? If economists want to be right, they have to say “Yes, it will, provided one understands that by “multiplier” one only means that there exists a ratio between money supply and monetary base.”

    What we actually need to do is start with the concept called “the neutrality of money,” which underlies almost all of macroeconomics, and has done so for hundreds of years.

    This concept is false. It rests on the fallacious notion that inflation increases everyone’s bank balances at the same time, at the same rate. In reality of course, inflation enters the economy at certain nodal points only, and as such, since each individual is responsible for a particular means of production in the overall division of labor, it means that inflation will in fact affect the real structure of the economy, even in the short run.

    It would be like me printing off $100 billion in my basement, and then spending it. I of course won’t spend that money on “GDP”, I won’t be increasing every individual’s bank balance at the same rate, but I will be increasing the bank balances of a certain group of people, and because of that, I will be bringing about a change in the structure of production.

    The neutrality of money assumption is a fallacious assumption. Money inflation is non-neutral in both the short run and, because the long run is determined by past short runs, it is non-neutral in the long run as well.

    The real economy was forever altered because of the fact that the Fed bailed out the banks in 2008, the same way the real economy was forever altered by the Great Depression, and even the Trojan War.

    So according to the neutrality of money, a 10% rise in the base will increase all nominal aggregates by 10% in the long run. Since M1 and M2 are nominal aggregates, they are affected just like nominal toaster expenditures and nominal Brazilian wax expenditures. (I keep mentioning the latter in a pathetic attempt to show that I am a hip 56 year old keeping up with new industries that didn’t exist when I was 20 years old.)

    Nominal aggregates are nothing but the sum total of nominal individual components. Of COURSE printing money will affect all aggregates that contain individual money units as components! It would be like saying “yelling” is supposedly neutral, because any increase in an individual’s yelling, will increase the nominal aggregate GDY (gross domestic yelling).

    But all this doesn’t mean that money is neutral. A 10% rise in the monetary base will not increase everyone cash balances at the same rate, as has been perfectly clear with the 2008 base increase. The base increased by a multiple factor, and yet every individual’s bank account did not increase at the same rate. The banks and the Treasury got lots of money. As a result, an economy where some people (banks, Treasury, etc) get more money from the Fed, whereas everyone else does not, their spending brings about a revolution in the structure of production. It affects real output, even in the short run.

    Of course wages and prices are sticky in the short run, so a 10% rise in the base does not have a proportional effect on most nominal aggregates in the short run. The neutrality of money (and the money multiplier) are long run propositions. However some economists believe the money multiplier is relatively stable when interest rates are positive, even in the short run. This is because the multiplier has two behavioral components; the reserve ratio and the currency/deposit ratio. Many economists think that during normal times (when interest rates are positive), banks hold very little in the way of excess reserves. Hence the reserve ratio is stable, and equal to the required reserve ratio. They also believe the currency/deposit ratio is relatively stable in the short run.

    It’s truly incredible how blind people can be when they think about prices, inflation, and the market process.

    To say that wages and prices are “sticky” in the short run, in the presence of inflation, is like saying hearing is “sticky” in the short run, in the presence of a loud explosion 10 miles away. Inflation is new money that enters the economy at distinct points. Of COURSE prices would be “sticky” in relation to such inflation, for the exact same reason that it takes time before your ears receive sound from a distant explosion.

    If banks receive a huge increase in reserves from the Fed, then it takes time for that money to be spent and respent throughout the economy such that prices in general rise. It’s a physical limitation of prices to not rise until people actually receive the additional money with which to spend!

    Saying that prices are “sticky” in the presence of inflation is like saying it’s possible that people can pay higher prices for goods before they receive more money with which to buy those goods.

    It would be like saying people’s hearing is “sticky” because they don’t instantly hear an explosion 10 miles away at the exact time the explosion occurred. No, people’s hearing is not “sticky”. What is “sticky” is the speed of sound.

    You say prices are “sticky”, but what you should be saying is that inflation spreading throughout the economy is “sticky.” People are physically unable to pay higher prices for the goods they buy if they haven’t even gotten the additional money yet. All this yammering of sticky prices is misguided. It is precisely inflation in our monetary system that is “sticky.” It is sticky because inflation enters the economy at distinct points, and the only way additional money can raise everyone’s cash balances is through making exchanges with each other using more money.

    The initial receivers first get the new money, which raises their cash balances first, then those initial receivers then spend more money, which then raises the cash balances of the next group of people, and then those people spend more money, and so on. As this occurs, prices rise in a one by one type fashion, some prices rise first, then other prices as money is spread throughout the economy and raises more and more people’s cash balances.

    Saying prices are “sticky” leads to the mistaken belief that people SHOULD raise prices sooner, before they even get the additional money, which is of course impossible.

    Prices, in a free market, are not “sticky.” It’s the ability of inflation to raise prices that is “sticky.”

    Inflation could raise all prices more or less at the same time, if only the Fed sent out checks to every individual who owns and deals in dollars. THEN you will see prices rise sooner. THEN you will realize it wasn’t prices that were sticky, it was the way inflation is carried out in our primary dealer system.

    This leads to the question of whether the money multiplier is useful. I don’t think it is, but let me try to explain why others disagree. They think a 10% rise in the base (when interest rates are positive) will result in a roughly 10% rise in M2, even in the short run. I have doubts, but I’m willing to accept that claim. Then they claim that changes in M2 have an important causal impact on NGDP—M2 is an important part of the monetary transmission mechanism. That’s the part I reject. I think future expected NGDP, plus asset prices, are the key transmission mechanisms. And future expected NGDP is driven by future expected changes in the supply and demand for base money. I don’t see anything special about M2.

    There is no constancy relation between base money and aggregate money stocks like M2, because there are no constancies in human action, which is responsible for those statistics. Period. End of story. Humans are not robots.

    All economists believe the money multiplier becomes highly unstable at zero rates, mostly because, well . . . because it does become highly unstable at zero rates. Yes, sometimes economists actually do believe what they see with their own eyes. I should add that if IOR is permanent, which seems increasingly likely, the money multiplier may become highly unstable at positive interest rates as well. Money will still be neutral in the long run, but the usefulness of the money multiplier (which I doubt even without IOR) will become even smaller. It will be as unimportant as the toaster multiplier, or the . . . Brazilian wax industry multiplier.

    Money is not neutral in the long run. Money is not neutral in the short run. Money entering the economy at distinct points makes neutral money an impossibility.

    Why oh why do so many economists believe in the myth of neutral money? Is it to dishonestly defend inflation against criticism? Is it just a matter of ignorance? Is it misunderstanding the nature of our monetary system? Is it believing money to be something it is not? It’s incredible to see so many economists make such a huge error over something so important.

  3. Gravatar of Mark A. Sadowski Mark A. Sadowski
    8. April 2012 at 19:30

    Scott wrote:
    “I should add that if IOR is permanent, which seems increasingly likely, the money multiplier may become highly unstable at positive interest rates as well.”

    Yes, I agree. Consequently I’ve come to the conclusion that the world’s major central banks have gone collectively mad.

    Correct me if I’m wrong, but by my count nine major central banks now pay IOR:

    1) RBA
    2) BOC
    3) BOE
    4) EBC
    5) BOJ
    6) RBNZ
    7) Norges Bank
    8) Riksbank
    9) FRB

    Based on the evidence it seems to do no major harm provided you’re not in ZIRP. Unfortunately the four most important banks on this list are essentially practicing ZIRP right now.

    And I still don’t understand the rational for IOR though God knows I’ve tried. The Fed’s stated justification seems to boil down to:

    1) Paying IOER helps to establish a lower bound on the federal funds rate.
    2) The payment of IOER permits the Federal Reserve to expand its balance sheet as necessary to provide the liquidity necessary to support financial stability.

    And this is helpful exactly *how* in ZIRP?

  4. Gravatar of OhMy OhMy
    8. April 2012 at 19:37

    There is no “multiplier”, there is a “divider” – banks make loans first and then convert some of their assets to reserves in the volume that is a fraction of the loans they had made. Most deposits are not reservable anyway.

    Empirics: loans precede deposits and reserves, so the “multiplier” name suggests a causality that is not there.
    http://www.federalreserve.gov/pubs/feds/2010/201041/201041pap.pdf

  5. Gravatar of Brito Brito
    8. April 2012 at 19:48

    OhMy, the system only works that way because the central bank allows it to do so by responding to demand for reserves; it doesn’t have to work that way.

  6. Gravatar of Dan Kervick Dan Kervick
    8. April 2012 at 20:07

    It is customary, I thought, to distinguish between the actual money multiplier – which is the actual ratio between reservable deposits and central bank reserves – and one of several theoretical money multipliers – the most commonly cited being the reciprocal of the reserve ratio. When policy makers use a money multiplier to estimate the impact of increased bank reserves on bank lending and deposits, they employ some theoretical multiplier to make the prediction.

  7. Gravatar of Doc Merlin Doc Merlin
    8. April 2012 at 22:31

    ‘“What we actually need to do is start with the concept called “the neutrality of money,” which underlies almost all of macroeconomics, and has done so for hundreds of years”’

    No, it hasn’t. You need to stop reading modern economists about this and actually read what people thought.

  8. Gravatar of Doc Merlin Doc Merlin
    8. April 2012 at 22:50

    “This concept is false. It rests on the fallacious notion that inflation increases everyone’s bank balances at the same time, at the same rate. In reality of course, inflation enters the economy at certain nodal points only, and as such, since each individual is responsible for a particular means of production in the overall division of labor, it means that inflation will in fact affect the real structure of the economy, even in the short run.”

    This couldn’t have been said better. Scott will object and say either (1) these effects are secondary or (2) that rational expectations will remove this effect.

    I disagree with both of these possible objections.

  9. Gravatar of david david
    9. April 2012 at 00:14

    Does neo-Austrian economics from Hayek to Boettke really disagree with the notion of a long-run neutrality of money?

  10. Gravatar of 123 123
    9. April 2012 at 00:50

    “I should add that if IOR is permanent, which seems increasingly likely, the money multiplier may become highly unstable at positive interest rates as well.”

    Yes, IOR is permanent in most large countries. The intellectual victory of Milton Friedman (’59) is impressive. See page 49 (p9 in a pdf) here:
    http://www.ny.frb.org/research/EPR/08v14n2/0809keis.pdf

  11. Gravatar of Steve Steve
    9. April 2012 at 01:17

    “Does neo-Austrian economics from Hayek to Boettke really disagree with the notion of a long-run neutrality of money?”

    Does Austrian economics make any testable predictions at all? Or is the term “Austrian economics” an oxymoron?

  12. Gravatar of Joe Joe
    9. April 2012 at 04:20

    I recall Hendrickson calling the multiplier as the demand for base money, as opposed to V which is the demand for the monetary aggregate. Now, expectations should temporarily change V, should they also temporarily change the multiplier? NGDP targeting should both affect the multiplier and V?

  13. Gravatar of OhMy OhMy
    9. April 2012 at 04:54

    Brito,

    Yes it does have to work that way, that is the point.

    The central bank can’t not supply reserves without destroying the payments system, losing control of the policy rate or both.

  14. Gravatar of Major_Freedom Major_Freedom
    9. April 2012 at 05:05

    Doc Merlin:

    This couldn’t have been said better. Scott will object and say either (1) these effects are secondary or (2) that rational expectations will remove this effect.

    I disagree with both of these possible objections.

    As do I.

    The effects, properly understood, characterizes the business cycle of boom / bust, and are hence not secondary.

    Rational expectations cannot possibly transcend the physical constraints that accompany monetary inflation into certain points of the economy, nor can it transcend the impossibility of anyone knowing where people will spend the new money they receive, before they spend it.

  15. Gravatar of Major_Freedom Major_Freedom
    9. April 2012 at 05:16

    Steve:

    Does Austrian economics make any testable predictions at all? Or is the term “Austrian economics” an oxymoron?

    Well, since you clearly believe economics is akin to astrology, in that you believe you can predict what people will learn in the future, and what they will choose to do in the future, before they themselves learn and do it, then you’re right, saying “Austrian economics” would be an oxymoron, since Austrianism isn’t astrology.

    Making testable predictions in human learning and action fallaciously presumes that there is constancy relations in human learning and action. In reality of course there are none, which is why economists tend not to be millionaires and billionaires, and are instead almost universally characterized by toiling away in sterile offices with no windows, making far less than entrepreneurs and businessmen, while they scribble on paper trying to find a magical equation that can enable them to predict what people will learn and do, before they learn and do.

    Making testable predictions is valid for subject matter that doesn’t learn and doesn’t act, like atoms, molecules, and electromagnetic waves. It is not valid for subject matter that learns from the very testing and predicting process itself and thus constantly transcends it.

    Einstein could not have predicted himself discovering the general theory of relativity, before he discovered it. He can only reconstruct his learning and discovery after the event.

    Each individual cannot predict his own learning path, and yet you believe that economists, who are individuals, can predict other people’s learning paths? It’s a chimera. A fool’s quest.

  16. Gravatar of Morgan Warstler Morgan Warstler
    9. April 2012 at 05:24

    Scott is only right in the long run.

    Where the money enters is DEFINITELY a problem.

    Scott’s response is that ANYONE can sell T-Bills to the Fed when it is making new money.

    The reality is that GS is doing that at favorable terms and likely knows in advance when Fed QE is coming.

    Question to Doc Merlin:

    what if the Fed bought gold? Drove up the price directly and paid off the Fox News gold bugs with their injections.

    THEN the new money doesn’t go to reserves, instead it is a bunch of anti-government folks terrified of inflation, they’d get rid of the new cash immediately buying guns, camo, food stuffs AND Brazilian waxes for their their girlfriends.

  17. Gravatar of Morgan Warstler Morgan Warstler
    9. April 2012 at 05:26

    “In reality of course there are none, which is why economists tend not to be millionaires and billionaires, and are instead almost universally characterized by toiling away in sterile offices with no windows, making far less than entrepreneurs and businessmen, while they scribble on paper trying to find a magical equation that can enable them to predict what people will learn and do, before they learn and do.”

    This is the agency problem.

    Somehow economics must return to studying the genius of entrepreneurs and businessmen.

  18. Gravatar of Dan Kervick Dan Kervick
    9. April 2012 at 05:27

    How many people have expectations about the supply and demand for base money?

  19. Gravatar of Morgan Warstler Morgan Warstler
    9. April 2012 at 05:33

    This is Waldman having never heard Scott discuss how NGDPLT pisses on booms:

    “On its own, NGDP path targeting would help “mop up” after financial fragility and collapse, because it weds depressions to inflations, engineering wealth transfers from creditors to debtors when things go wrong. But we’d rather avoid the whole cycle of fragility, insolvency, and inflation, if we can.”

    I wonder why he hasn’t heard about it??

    Oh yeah, Scott never talks about it.

  20. Gravatar of bill woolsey bill woolsey
    9. April 2012 at 07:13

    I think that Austrian economics is consistent with the long run neutrality of money. The short run nonneutrality generates the malinvestments, but they “must be” malinvestments because of long run neutrality.

    The Mises theory of the crack-up boom, where every higher inflation is needed to maintain malinvestments, resulting in the collapse of the monetary economy implies superneutrality as well.

  21. Gravatar of bill woolsey bill woolsey
    9. April 2012 at 07:38

    I support interest on reserves.

    The interest rate on reserve balances should be pegged below the T-bill rate.

    That way, changes in the interest rate on short and safe assets will have little or no impact on the demand for reserves.

  22. Gravatar of Peter N Peter N
    9. April 2012 at 08:08

    @Major_Freedom

    There are a few other reasons for stickyness.

    1) Contractual Stickiness. A contract specifying a price or pay rate in nominal money for some period is sticky for that period. This is true for wages, prices, debt and, to a large extent, taxes.

    2) the twins – Monopolies and Lack of pricing power

    3) Opportunity costs.

    I believe that 1 is the most important, and it amazes me that economists have to go search for what completely surrounds them. Some of them must have mortgages or employment contracts.

  23. Gravatar of Major_Freedom Major_Freedom
    9. April 2012 at 09:59

    Morgan Warstler:

    This is the agency problem.

    Somehow economics must return to studying the genius of entrepreneurs and businessmen.

    Only an economics that is based on the individual, not aggregates, can do this.

    bill woolsey:

    I think that Austrian economics is consistent with the long run neutrality of money. The short run nonneutrality generates the malinvestments, but they “must be” malinvestments because of long run neutrality.

    That doesn’t follow.

    The generation of malinvestments is such because of their unsustainability in the real sense as it pertains to real consumer preference. This does not require long term neutrality of money because even when the malinvestments are identified, corrected, and people move on with their lives, the very sequence of events in that boom bust cycle, even the corrections, forever alters the structure of production away from what it would have otherwise have been absent the inflation. The corrections can take place, but it doesn’t put the economy back onto the same exact track as would have otherwise existed without the inflation.

    Humans always utilize what they have when engaging in economic behavior at any given time. And what they have to work with in the present is always a cumulative sum of all past economic events, including booms and busts, including corrections.

    Does this mean that even the notion of “correcting” malinvestments is the wrong word, if no corrections could ever put the economy on the same track as the one where inflation never took place? It may seem to be so, but it actually isn’t, no more than someone gaining a whole bunch of unhealthy weight, can be said to be “correcting” their lifestyle by losing the weight, despite the fact that their life will never be the same compared to if they lived a life where they never gained the weight in the first place.

    Neither Rothbard nor Mises held that money is neutral. Austrians in general reject the neutrality of money.

    The easiest way to understand that Austrians hold money to not be neutral even in the long run, is to simply realize that they believe people would have been better off in the long run if the boom bust cycle never occurred. That is enough to understand they believe money is not neutral in the long run.

    The Mises theory of the crack-up boom, where every higher inflation is needed to maintain malinvestments, resulting in the collapse of the monetary economy implies superneutrality as well.

    Not even close. The super-neutrality of money requires that changes in the growth rate of the money supply exert no effects on output. Clearly, CLEARLY, if higher inflation is needed to maintain malinvestments, resulting in a collapse of the monetary system, carries with it a HUGE effect on the real economy. It is precisely because money manipulation is capable of ruining an economy that is proof positive of the non-super-neutrality of money.

    Austrians hold that money is non-neutral for logical reasons. According to a praxeological analysis, the axiom of human action represents an irrefutable truth.

    The law of diminishing marginal utility is logically implied in the axiom of human action — and thus irrefutably true. Money, since it is an economic good, is subject to the law of diminishing marginal utility. Therefore, a rise in the money supply must also necessarily lead to a decline in its purchasing power and must cause changes in other economic variables.

  24. Gravatar of david david
    9. April 2012 at 10:09

    I would pay good money for anybody to explain how these two statements could be true at the same time:

    Making testable predictions in human learning and action fallaciously presumes that there is constancy relations in human learning and action. In reality of course there are none…

    The super-neutrality of money requires that changes in the growth rate of the money supply exert no effects on output. …

    Austrians hold that money is non-neutral for logical reasons. According to a praxeological analysis, the axiom of human action represents an irrefutable truth.

    The law of diminishing marginal utility is logically implied in the axiom of human action — and thus irrefutably true.

    Behold, assumptions on human behavior which both imply nothing testable about reality, yet tell us about the long-run neutrality of money…? ?????

  25. Gravatar of Major_Freedom Major_Freedom
    9. April 2012 at 10:09

    Peter N:

    There are a few other reasons for stickyness.

    1) Contractual Stickiness. A contract specifying a price or pay rate in nominal money for some period is sticky for that period. This is true for wages, prices, debt and, to a large extent, taxes.

    Sticky compared to what exactly? Compared to contracts that would see wages and incomes and taxes changing to the penny, every second of every day, depending on instantaneous changes in supply and demand for all goods and services all across the world that occur each second?

    Using a real world standard, prices and wages are maximally flexible in a free market, even more flexible than in a world of central banks preventing large movements in prices and spending via inflation.

    You’re not talking about sticky prices, you’re talking about sticky information spreading throughout the economy over time as opposed to instantly informing people as to the exact up to date information concerning supply and demand for all goods and services, every second of every day.

    2) the twins – Monopolies and Lack of pricing power

    You mean the central bank monopoly? The lack of purchasing power of money (pricing) power among non-central bank agents?

    3) Opportunity costs.

    Opportunity costs are ubiquitous and logically inherent in all economic action open to choice, even inflation of the money supply. Inflation therefore cannot possibly be presented a solution to this supposed cause of price stickiness.

    I believe that 1 is the most important, and it amazes me that economists have to go search for what completely surrounds them. Some of them must have mortgages or employment contracts.

    Speaking of contract stickiness, isn’t the Fed’s charter an example of contract stickiness? They are contractually obligated to a “dual mandate”.

  26. Gravatar of ssumner ssumner
    9. April 2012 at 10:39

    Nathan, You said;

    “The argument is about the causality between variables in the money multiplier.”

    What argument?

    MF, You said;

    “The neutrality of money assumption is a fallacious assumption. Money inflation is non-neutral in both the short run and, because the long run is determined by past short runs, it is non-neutral in the long run as well.”

    I don’t know why other economists never thought of that.

    Mark, That’s what I wonder too.

    OhMy, Think about how the “toaster multiplier” works, that might cause you to rethink the money multiplier. Many commenters seem to be talking about the real side of the banking industry, when the money multiplier is fundamentally a nominal concept, an application of the neutrality of money. Hence if makes no difference whether causality runs from loans to deposits or from deposits to loans (both views are false, BTW; the two are determined jointly if banks maximize profits.)

    Dan, You said;

    “It is customary, I thought, to distinguish between the actual money multiplier – which is the actual ratio between reservable deposits and central bank reserves – and one of several theoretical money multipliers – the most commonly cited being the reciprocal of the reserve ratio. When policy makers use a money multiplier to estimate the impact of increased bank reserves on bank lending and deposits, they employ some theoretical multiplier to make the prediction.”

    Only if they are complete idiots. The vast majority of base injections go into currency, except when rates are near zero. And the reserve multiplier is nearly useless when rates are near zero.

    Doc Merlin, You said;

    “No, it hasn’t. You need to stop reading modern economists about this and actually read what people thought.”

    Actually I’ve read Hume, Thornton, Fisher, and lots of other non-modern types, and they also relied on the neutrality of money..

    You said;

    “Scott will object and say either (1) these effects are secondary or (2) that rational expectations will remove this effect.”

    I will claim that MF completely misread what I wrote, I never denied short run non-neutralities.

    David, I don’t know.

    123, Yes, a victory, but a tiny one. When rates are positive the level of reserves is tiny compared to currency–which still pays no interest.

    Joe, Yes, expectations affect both the money multiplier and V, by changing nominal interest rates.

    OhMy, You said;

    “The central bank can’t not supply reserves without destroying the payments system, losing control of the policy rate or both.”

    Of course it must lose control of the policy rate, if it didn’t we’d end up with hyperinflation or hyperdeflation.

    Morgan, You said;

    “Scott is only right in the long run.”

    Didn’t I say that money neutrality only holds in the long run?

    Dan Kervick, You said;

    “How many people have expectations about the supply and demand for base money?”

    Not independently, but certainly in combination. Or do you think people don’t have inflation expectations?

    Bill, That’s fine, but I’d point out that if you “favor” interest on reserves below the T-bill yield, readers should understand that you OPPOSE the current Fed policy of paying positive interest on reserves. I’m fine with that, but readers might have assumed you and I disagreed.

    David, I’ll let the Austrians handle that.

  27. Gravatar of Saturos Saturos
    9. April 2012 at 10:50

    Truly capitalism is glorious, for now one need not be bothered by body hair wheresoever one finds it. (I’m trying to wax lyrical about innovation like Don Boudreaux. But I guess it takes practice.)

    Bill, are you saying that ABC is compatible with Rational Expectiations? What does the neutrality of money (or otherwise) have to do with systematic, regular malinvestments?

    Scott, I caught your reference to misunderstanding velocity. You’re right, I’ve amended my old comment on the previous post.

  28. Gravatar of Morgan Warstler Morgan Warstler
    9. April 2012 at 11:15

    If we’d do my Guaranteed Income plan the cost of Brazilian waxes would fall dramatically.

    Truly dumbfounded at the resistance to such positive upside.

  29. Gravatar of Saturos Saturos
    9. April 2012 at 11:17

    “If we’d do my Guaranteed Income plan the cost of Brazilian waxes would fall dramatically.”

    Seriously, how would that work?

  30. Gravatar of 123 123
    9. April 2012 at 11:25

    @Scott
    @Bill Woolsey

    Scott:
    “That’s fine, but I’d point out that if you “favor” interest on reserves below the T-bill yield, readers should understand that you OPPOSE the current Fed policy of paying positive interest on reserves. I’m fine with that, but readers might have assumed you and I disagreed.”

    1. Bernanke (or at least someone inside the Fed) agrees. If the sterilized QE is implemented, the Fed would have a decent chance of paying the below T-bill yield on at least some of its liabilities.

    2. Fannie and Freddie receive zero IOR, Citi gets 25bps IOR, so it is not clear which IOR (Fannie’s or Citi’s) should be compared with the T-bill yield.

    3. Early January 3 month German T-bills yielded negative 10bps, so Draghi’s 25 bps deposit rate was too high according to your criteria.
    However, when Trichet was at his tightest, when he increased the deposit rate to 75 bps last July, 3 month German T-bills yielded 108 bps.
    I like Draghi’s policy of paying above T-bill IOR much more than I liked Trichet’s policy of paying below T-bill IOR.

  31. Gravatar of Nick Rowe Nick Rowe
    9. April 2012 at 14:39

    Scott: when you talk about the “reserve ratio” and “excess reserves” I think it matters how they are defined. I prefer to define them as the *desired* reserve ratio, not the *required* reserve ratio (which may not exist), and “excess” reserves being actual reserves minus *desired* rather than *required* reserves.

  32. Gravatar of OhMy OhMy
    9. April 2012 at 15:14

    SS,

    The direction does make a huge difference. You say “They think a 10% rise in the base (when interest rates are positive) will result in a roughly 10% rise in M2, even in the short run. I have doubts, but I’m willing to accept that claim.” In light of empirical evidence (see my link above) it makes more sense to say: “a 10% rise in M2 will lead to a roughly 10% rise in the base”. So if you are saying that both views on the direction of causality are “wrong” then your sentence above makes no sense.
    Your statement that “the two are determined jointly if banks maximize profits” is anyway false in the light of the paper I linked to.

  33. Gravatar of Russ Anderson Russ Anderson
    9. April 2012 at 18:29

    Scott wrote:

    ‘“What we actually need to do is start with the concept called “the neutrality of money,” which underlies almost all of macroeconomics, and has done so for hundreds of years”’

    Doc Merlin, replied:

    “No, it hasn’t. You need to stop reading modern economists about this and actually read what people thought.”

    Scott replied:

    “Actually I’ve read Hume, Thornton, Fisher, and lots of other non-modern types, and they also relied on the neutrality of money..”

    I’ve read lots of people that were wrong, too. Not everyone believes “the neutrality of money” theory, especially in the short run and some in the long run. Your first sentence does not seem to acknowledge that disagreement.

    http://modeledbehavior.com/2010/11/08/the-neutrality-of-money/

  34. Gravatar of Do you believe in the money multiplier? « Economics Info Do you believe in the money multiplier? « Economics Info
    9. April 2012 at 23:01

    [...] Source [...]

  35. Gravatar of Bill Woolsey Bill Woolsey
    10. April 2012 at 05:50

    Scott:

    If T-bill rates rise, and not too much, I would favor positive interest rates on reserves balances.

    I favor making reserve balances into a money market mutual fund type instrument, invested in T-bills with the a service charge in proportion to the size of the portfolio.

    Because I favor using the level of nominal GDP (rising over time) as the fundamental nominal value determined by the monetary order, the long run neutrality of a given stock of money is irrelevant. How interest paid on reserves might impact the long run neutrality of a given stock of noninterest bearing money just doesn’t matter.

    I think that the nominal quantity of money should adjust to meet the demand for it. As you can see here, I still am not entirely confortable saying that the yield on money can adjust as well to bring them into equilibrium.

    But, because the nominal demand for money depends on whatever fundamental nominal value is determined, then this rule that the nominal quantity of money (and the yield on it) adjust to keep the nominal amount demanded equal to the nominal uantity are subsidiary to the rule that the fundamental nominal value be determined.

    For example, suppose the dollar is defined as 1/20th of an ounce of gold, but every one uses paper money. The quantity of paper money (and its nominal yield if any) should adjust to that the nominal amount demanded and the nominal quantity are equal. But, subject to the side constraint that the money price of gold is $20.

    If the market price of gold is $22, and the nominal quantity of money is equal to the nominal amount demanded, that will hardly do.

    Of course, I don’t favor using gold. The side constraint is that nominal GDP (or really, its expected value) remain on target. Subject to that side constraint, the quantity of money should match the amount demanded.

  36. Gravatar of Major_Freedom Major_Freedom
    10. April 2012 at 07:40

    David:

    I would pay good money for anybody to explain how these two statements could be true at the same time:

    Making testable predictions in human learning and action fallaciously presumes that there is constancy relations in human learning and action. In reality of course there are none…

    The super-neutrality of money requires that changes in the growth rate of the money supply exert no effects on output. Austrians hold that money is non-neutral for logical reasons. According to a praxeological analysis, the axiom of human action represents an irrefutable truth. The law of diminishing marginal utility is logically implied in the axiom of human action — and thus irrefutably true.

    I would pay even more money for you to explain how these two statements could NOT be true at the same time. Is it the seeming conflict between “you can’t make predictions!” and “Money is non-neutral and WILL have an effect”? Is that it? Well that’s easily explained. I can’t predict that the Fed will choose to inflate in the future. But I can know the effects of inflation if they do.

    The logical necessities I am talking about are logical constraints, not empirical predictions of choices. I can’t know what you will eat next year at this time, but I can say that whatever you do eat, will be constrained to the principles of marginal utility.

    Does that help?

    Behold, assumptions on human behavior which both imply nothing testable about reality, yet tell us about the long-run neutrality of money…? ?????

    These are not merely assumptions, they are logically necessary. And they don’t tell us there is a long term neutrality of money, indeed the opposite.

  37. Gravatar of Major_Freedom Major_Freedom
    10. April 2012 at 07:52

    ssumner:

    “The neutrality of money assumption is a fallacious assumption. Money inflation is non-neutral in both the short run and, because the long run is determined by past short runs, it is non-neutral in the long run as well.”

    I don’t know why other economists never thought of that.

    Well, other economists have thought of it, one just has to know where to look. Most mainstream economists unfortunately relegate themselves to dealing with statistical aggregates only, and aggregates that are observable only. Their models cannot help but overlook the micro level effects of inflation, and they cannot help but overlook the required counter-factual analysis.

    Only economists who ultimately ground their economics in individual human action, where choices and intentions take center stage, can pick up these sorts of things.

  38. Gravatar of 123 123
    10. April 2012 at 09:10

    “Yes, a victory, but a tiny one. When rates are positive the level of reserves is tiny compared to currency–which still pays no interest.”

    No. Just looked up the 2005 balance sheet of Norges bank. Currency was 55% of M0.

    The victory was not complete, as Norway did not adopt Friedman’s proposal to privatize notes and coins.

  39. Gravatar of ssumner ssumner
    10. April 2012 at 10:41

    Morgan, Now you’re talking!

    123, Better yet would be Draghi’s lower interest rates, plus the IOR below the T-bill yield.

    Nick, I’m fine with that. But as long as rates are positive and there is no IOR, then the actual reserve ratio will be close to the legal reserve ratio (or close to zero if no such legal minimum exists.) There is actually no need to even talk about ERs.

    OhMy, You can play all sorts of “causality” games. If you view the 2% inflation target as exogenous, then it “causes” everything else (MB, M2, interest rates, loans, deposits, etc, are all endogenous.) I prefer to think in terms of the way the central bank achieves it’s inflation target, which is by adjusting the base.

    Russ, You said;

    “Not everyone believes “the neutrality of money” theory, especially in the short run”

    And who claimed they do? The fact is that the long run neutrality of money underlies both classical and modern mainstrean econ. Yes, not every single economist agrees, but that’s no surprise.

    Bill Woolsey, That’s certainly a reasonable system.

    123, I meant if rates are positive and there is no IOR. In other words the welfare loss Friedman worried about was tiny, because reserves were tiny when he made that proposal.

  40. Gravatar of 123 123
    10. April 2012 at 19:56

    “I meant if rates are positive and there is no IOR. In other words the welfare loss Friedman worried about was tiny, because reserves were tiny when he made that proposal.”

    No. According to Friedman, no IOR is equivalent to IOR plus a tax on reserves. So the deadweight loss is large when the basw is tiny.

    “Better yet would be Draghi’s lower interest rates, plus the IOR below the T-bill yield.”

    So 3 month German T-bill yield stays the same at Draghi levels, IOR is minus 20 bps, and quantity of base money is reduced by 500 billion? A nightmare.

  41. Gravatar of Morgan Warstler Morgan Warstler
    11. April 2012 at 09:10

    Saturos, glad you asked!

    Under my plan the 30M currently unemployed who register to work receive a Paypal Debit Card and each Friday night they have $240 deposited ($6 per hour).

    They also have to maintain a new Ebay CV page where their upcoming work week is auctioned off starting at $1 per hour (min bid $40 per week).

    It goes without saying, but at $1 per hour, everyone is worth something to someone. But, if someone is unbid on, they still get the $240. I expect 1M work from home telephone operators at the low low end. (This is a side benefit to American safety net, you always talk to a human.)

    They get to keep 50% of winning bid.

    They can decline certain kinds of work.

    They are not required to travel more than 5 miles for work.

    Winning bidders have pre-funded accts. and are not required to pay employment taxes etc. Insurance is sold on site for a couple bucks for those who need it.

    —–

    Things are kept fuzzy precisely to allow the web eco-system to take over.

    Finding workers who keep getting won by same bidder is easy, this is proof of value, bids rise accordingly.

    Ex: We see neighborhoods hiring for shared services and eyeballing work load. 30 neighbors go in for $4 per week of have their dogs cleaned up after. they bid $120. winner seizes deal he can finish work in 3 hours a day, or two days a week, for the extra $60 bucks.

    Local requirements mean that newcos built to use this low cost labor, establish themselves in depressed areas.

    Brazilian waxes offered at these locations are super cheap! Deal finders will drive to bad neighborhoods to get waxed.

    Current waxers will lower prices a bit to compete.

    win-win-win

  42. Gravatar of david stinson david stinson
    11. April 2012 at 11:44

    Hi Scott.

    I hope I am not too late.

    Further to MF’s point, when macroeconomists speak of the short run and the long run, are they referring to them in the dynamic sense – i.e., the passage of real world time – or are they referring to them in a quasi-microeconomic sense? In micro, the concepts of short and long run are meant to capture not the passage of time per se but an interval of time – i.e., the long run does not “follow” the short run. In micro, the longer interval translates into a greater degree of flexibility over the means of production (you can make more decisions simultaneously). I am thinking that the analogous concept in macro might vary depending on the school – examples might be a) sticky vs perfectly flexible prices, b) discoordinated vs coordinated expectations, or whatever.

  43. Gravatar of Long-Run Neutrality | Economic Thought Long-Run Neutrality | Economic Thought
    12. April 2012 at 08:29

    [...] is a a minor debate in the comments section of a Scott Sumner post on long-run money neutrality.  The substance of the “debate” is not that great, but [...]

  44. Gravatar of ssumner ssumner
    12. April 2012 at 11:36

    123, Why assume a lower base money?

    Your point about the deadweight loss was valid, but given plausible elasticities I still can’t see it as a big deal.

    David, Yes, it’s like in micro–in the long run wages and prices adjust.

  45. Gravatar of Major_Freedom Major_Freedom
    12. April 2012 at 11:50

    David, Yes, it’s like in micro–in the long run wages and prices adjust.

    Not if there is yet another short term deflation/inflation at the time that requires more wage and price adjustments.

    In an inflationary economy, there is constant adjusting of wages and prices.

    It’s like saying an initial gust of wind against a sailboat has long run effects on the sailboat’s trajectory, as if the subsequent gusts of wind don’t even exist and have no effect.

  46. Gravatar of 123 123
    12. April 2012 at 13:00

    “Why assume a lower base money?”
    This is the way you make it scarce enough. Otherwise T-bills are more attractive and yield less than IOR.

    “Your point about the deadweight loss was valid, but given plausible elasticities I still can’t see it as a big deal.”
    Deadweight loss is tiny when close substitutes are readily available.
    However, during financial panic substitutes are scarce, and deadweight loss explodes. That’s why Bernanke was right to start paying IOR.

  47. Gravatar of flow5 flow5
    15. April 2012 at 19:05

    “it actually has nothing to do with the velocity of money”

    If you’re talking about Milton Friedman’s “monetary base”, then that’s not true.

    “rise in the base does not have a proportional effect on most nominal aggregates in the short run”

    When legal reserves were binding (& commercial banks held few excess reserves of significance), there was always an immediate expansion of commercial bank credit (between 1942 & 1995), given any injection of Reserve Bank credit (supplying additional excess reserves). I.e., the commercial banking system’s expansion coefficient was always stable (even in the very short run).

    “leads to the question of whether the money multiplier is useful”

    Legal reserves are no longer “binding” only in the sense they don’t prevent CBs from lending & investing. But that’s not true when the FED drains Reserve Bank Credit, & the CBs contract credit in response.

Leave a Reply