Trouble with the equation of exchange.

[itfitzme]

Yeah see I have a pure maths background, so I approaching things with "is this a helpful way to think about something; does it yield any interesting results?" rather than trying to map concepts to some physical representations.

So the equation of exchange is just an identity. We can't get anything useful out of an identity. First off let's write it in terms of growth rates rather than levels, so MV=PY => %M + %V = %P + %Y. We have to add some rules to make it interesting. The first and most obvious being that Y returns to its natural rate in the long run, it's long run growth rate being the trend growth in productivity + trend growth in population. Say about 3%. Next we ask what measure of money we should pick for M. Friedman liked to use M2 because M2 did something cool. I can't find it now, but I recall seeing Friedman showing that the long run demand for money when money is measured as M2 is stable. So the velocity of M2 is stable, hence %V = 0. So the only things left to vary are M and P, which gives you the conclusion that inflation (%P) is everywhere and always a monetary phenomenon.

This turns the "equation of exchange", a useless identity, into "the quantity theory of money". An immediate consequence of which is that inflation = %Y - %M = 3% - %M. So if you want to have zero inflation, you should target M2 growth of 3% (Friedman's K-percent rule). On top of that, if there's a stable relationship between "how much" monetary base gets "turned into" M2, you can achieve stable prices with a constant monetary base rule.

Of course the problem is that over the short run velocity isn't stable (in the same way the exchange rate isn't stable around PPP in the short run), and the economy can be hit by productivity shocks which change the natural rate of ouput. And of course the way MB affects M2 is volatile.

So over the short run the quantity theory of money is useless. The reason you'll see me citing MV=PY all over the place is because people frequently misapply QTM thinking that velocity is constant over the short run. So you get shit like "interest rates rates were low, that means easy money!".

And yet, in things like electrodynamics identities lead to useful deductions. Interestingly, the history of electrical engineering, that was added to by Farady, Henry, Volta, and others, was then picked up by Maxwell who began with identities and deduced that the speed of light is dependent on only the permeability and permittivity of free space. It is an interesting process of taking a bunch of mathematical identities that were tested under the condition that the mathematical representations map into physical quantities. The combination of a few identities combined to show that c=root(e0*u0). Every thing else, relative speed, etc, just canceled out. That result was picked up by Einstein who then leveraged it to produce his theory of special relativity.

Of course, the history of modern physics seems to be an interchange between the theoretical physics, more pure mathematics types, revealing some things and the applied field reveling other. Each often getting the other unstuck. Currently, they are working on the Higgs boson where the the pure math is able to suggest where to look and the particle physicists are now looking.

In MV=PQ, P and q are identifiable quantities. M is a countable quantity and V can, in a simpler system, be identified. But, each are physically different objects. As such, while mathematically it is simply an identity, it shows how these otherwise different observable phenomena are in fact related.

It's similar to voltage and current which are measurable and physically different quantities. The identity, V = RI is simply an identity. But it becomes pretty important when combined with the identities related to other objects that are in the same physical system. V=c*dI/dt, and I=L*dV/dt are also simply identities. But combined in the same physical system, the I becomes a common element of flow and suddenly, the performance of all of the quantities become interdependent.

I am in no disagreement with the idea that the frequency of money is not necessarily constant. I have yet to see any way to deduce or demonstrate even it's upper and lower boundaries.

It will take me a while to absorb your comments. Unfortunately, I am afraid you haven't presented it in a manner that I am going to find easy to recognize. But I'll give it a shot starting with rewriting it in terms of growth rates and following through. It might be helpful if you can show how

MV=PY => %M + %V = %P + %Y

is arrived at.

And I will see if i can find Friedman and his M2.

Being that V and M are not directly measurable, just too many people, it will be nice to see how Friedmen got around this issue.

And M2 is nice in that it doesn't include reserves. It includes the stock of money available to enter the flow, demand deposits and the like. But, it isn't the stock in the flow. For all I know, this is why it is considered that V is not constant. Without either a direct measure of it, a solid deduction based on underlying physical quantities, or some valid deduction based on an alternative and measurable quantity, it remains simply unknown. That becomes a real problem. All I find so far are statements that it isn't constant, without any proof, or s presentation of VM1, which isn't M or V.

And it is a fact to say that V is not infinite nor can it increase without bound. It does rely on physical processes that limit it. One dollar can only go round the loop so fast, even if it is just two people handing the same dollar back and forth between them. It may very well be that the introduction of electronic transactions have substantially increased the rate.

M2 may serve as an indirect measure of M. I am not sure what we mean by M2 is stable. M1 and M2 have increase constantly since 1959. I haven't converted that to real or per capita.

The question remains, how does M2 actually increase. It certainly has, as has M1. But, it does not account for all money in the flow, especially if the flow is dependent on some other source or sink. There is no doubt that the change of M1 and M2 are connected to the change of M. They are sources and sinks for M. And, if in fact, no other method exist for putting debt free money into the flow, there is only one way that these could increase. All physics models are based on capturing all sources or sinks. Otherwise, you have some unknown that may be serving as a buffer that has a performance which is correlated to the level of the flow, but isn't it. So we look at M1 and M2, increasing yet have no defined source for that increase.

Ergo, why I am looking at if from the perspective of

(Mf + dC + dS) V = PQ = GDP(2)

because dS captures the savings, demand, and other checkable deposits as they serve as a source and sink to the flow. dS is the change in M1 and M2. Clearly, if everyone keeps $200 in their savings account, thus elevating the M1 and M2, but is not used for M which accounts for P, then doesn't to do anything. And, unlike simpler processes where the level of the source does have a definable effect on it's rate of change, I don't see we can say the same for savings.

Let's face it, it's just money and transactions, it's just not that complicated. The complicated part is that its based on 7 million businesses, 160 million workers, and twice that in population. That makes it difficult to observe M and V. P and Q we can get by survey or sample.

I am left, there for, to conclude that (Mf + dC + dS) V = PQ = GDP(2), given that V cannot increase unbounded and no mechanism exists to increase Mf, leads to the unmistakable conclusion that dC, credit, remains the only source of money to increase the flow, in the long run.

 

[DSGE]

It will take me a while to absorb your comments. Unfortunately, I am afraid you haven't presented it in a manner that I am going to find easy to recognize. But I'll give it a shot starting with rewriting it in terms of growth rates and following through. It might be helpful if you can show how

MV=PY => %M + %V = %P + %Y

is arrived at.

Sure. So a cool little trick that happens all over the place in economics that they never tell us about in maths. It's easier to just write it out so:

Being that V and M are not directly measurable, just too many people, it will be nice to see how Friedman got around this issue.

Well M is directly measurable. M is just some money stock. But he did it, if I recall, by coming up with a money demand function which was a function of permanent income and relative rates of return on all sorts of assets, and found that money demand is only sensitive to permanent income, which is relatively stable. I wish I could find the damn thing...

And M2 is nice in that it doesn't include reserves.

Not sure why reserves is an issue...

It includes the stock of money available to enter the flow, demand deposits and the like. But, it isn't the stock in the flow.

Because M isn't a "stock in the flow" whatever that means. It's just a stock. V tells you about the flow. So if you take the monetary base, for example, and if banks start holding excess reserves, that shows up as a fall in V.

When you say "the stock in the flow" you mean the quantity of money that is changing hands (being used for exchange). If you add up all that, that's just NGDP! With that definition of M the equation just becomes M = PY. What we do is take any old stock of money we choose, M, and V tells us how much of it is flowing.

So we need to choose an M to talk about that actually gives us some interesting results. Take the monetary base again: Theory tells us that in a liquidity trap, bank's demand for money will become perfectly elastic. So any increase in M will result in an equal decrease in V. That's an interesting result. Because it means no matter how much we expand the monetary base, NGDP won't budge (we won't get inflation). [Of course that statement of the theory is lacking in the expectations area, but that's neither here nor there].

It doesn't make sense to talk about "the stock in the flow".

M2 may serve as an indirect measure of M. I am not sure what we mean by M2 is stable.

Not "M2 is stable", the "demand for M2" is stable. Ie, V doesn't change very much. So we can set %V = 0.

The question remains, how does M2 actually increase. It certainly has, as has M1. But, it does not account for all money in the flow, especially if the flow is dependent on some other source or sink. There is no doubt that the change of M1 and M2 are connected to the change of M. They are sources and sinks for M. And, if in fact, no other method exist for putting debt free money into the flow, there is only one way that these could increase. All physics models are based on capturing all sources or sinks. Otherwise, you have some unknown that may be serving as a buffer that has a performance which is correlated to the level of the flow, but isn't it. So we look at M1 and M2, increasing yet have no defined source for that increase.

Ergo, why I am looking at if from the perspective of

(Mf + dC + dS) V = PQ = GDP(2)

because dS captures the savings, demand, and other checkable deposits as they serve as a source and sink to the flow. dS is the change in M1 and M2. Clearly, if everyone keeps $200 in their savings account, thus elevating the M1 and M2, but is not used for M which accounts for P, then doesn't to do anything. And, unlike simpler processes where the level of the source does have a definable effect on it's rate of change, I don't see we can say the same for savings.

Let's face it, it's just money and transactions, it's just not that complicated. The complicated part is that its based on 7 million businesses, 160 million workers, and twice that in population. That makes it difficult to observe M and V. P and Q we can get by survey or sample.

I am left, there for, to conclude that (Mf + dC + dS) V = PQ = GDP(2), given that V cannot increase unbounded and no mechanism exists to increase Mf, leads to the unmistakable conclusion that dC, credit, remains the only source of money to increase the flow, in the long run.

Maybe we should continue with that stuff once we deal with the "what is M" question. But yes, even if we fix the monetary base (something Friedman wanted to do toward the end), M1 and M2 would still grow endogenously through an increase in credit.

 

[EdwardBaiamonte]

I do like the idea of using inflation to get the money out from under the matresses on those that have amassed huge quantities of it. The engine of the economy functions on the flow of money, not it sitting static. I am concerned that the past few decades of inflation have been offset by adding debt.

but, lets not forget that when you appoint a committee of liberal geniuses to inflate the currency they will almost certainly guess incorrectly just as Hitler and Napoleon did when they invaded Russia or when Stalin and Mao did when they ran their economies. The business of real growth cant really be jump started since you will cause too many wasteful false starts.

The only real solution is a broadly based gold standard since it will produce the most stability and so the best foundation on which real growth can take place. Having a counter culture left wing unstable nut-job like BO in the White House is the last thing you want in a crisis. His ideology is a crisis of instability all by itself.

 

[itfitzme]

Maybe we should continue with that stuff once we deal with the "what is M" question. But yes, even if we fix the monetary base (something Friedman wanted to do toward the end), M1 and M2 would still grow endogenously through an increase in credit.

Thankfully, they do present "tricks" when doing the maths in physics and engineering. Every so often a prof would use some "trick", without pointing it out, and we'd all be scratching our heads wondering how he came up with that one.

I'm back to the data now which gives me the host of real delta whatevers.

And one man's savings is another man's debt, which adds to the mess.

And then a problem is that credit, money supplies, population, savings, CPI, GDP all increase a bit exponentially. That CPI and population both increase exponentially doesn't make them really correlated. So I have to make sure that I prove the physical connections.

It would be great if I could find the history of Fed monetary policy activities, in numbers. How much have the sold and purchased during open market operations? By how much and when did they increase reserves?

I'm sitting here, staring at

It shows the change in consumer credit bracketed by the recession.

It is more apparent when smoothed and the rate of change plotted

The problem being that the resolution is to course to demonstrate causality. And, smoothing decreases information.

Then there is

Which shows "savings" (M2-M1) and consumer credit in real terms per capita.

The correlation is apparent but the lead and lag changes. So, as interesting as it is to look at, there are other factors involved.

And, of course, even expectation becomes significant. In a totally unrelated incident, the debt crisis and the S&P downgrade of the US, the DJI shows not just the obvious drop, but an increase in variance over the following months.

It reacted exactly like we expect of an under damped control system. The variance increases after the event and slowly dampens out, but not by much. That is, of course, a purely psychological effect.

So all in all, there is all sorts of things going on, from the very broad, to the very detailed.

 

[Norman]

MV=ΣPQ

The equation in this form actually tells us nothing about real world. It can never tell 1 dollar = 1 bread. The sides most always be the same.

Thus it can't be used to predict inflation. Since you actually need real goods to measure inflation. This equation happens entirely in the nominal world. Nominal GDP = Nominal GDP measured in 2 different ways. Velocity of money * money stock = all the money spent = NGDP and Average price * goods sold = all the money spent = NGDP. As you can see you can't tell if price of bread or prices in general increased using that, for that you do need to live in real world. NGDP =! real gdp.

I think the equation is a bit unnecessary when there is already much better way to understand price system - supply and demand.

 

[DSGE]

MV=ΣPQ

The equation in this form actually tells us nothing about real world. It can never tell 1 dollar = 1 bread. The sides most always be the same.

Thus it can't be used to predict inflation. Since you actually need real goods to measure inflation. This equation happens entirely in the nominal world. Nominal GDP = Nominal GDP measured in 2 different ways. Velocity of money * money stock = all the money spent = NGDP and Average price * goods sold = all the money spent = NGDP. As you can see you can't tell if price of bread or prices in general increased using that, for that you do need to live in real world. NGDP =! real gdp.

I think the equation is a bit unnecessary when there is already much better way to understand price system - supply and demand.

Yeah, you have to add things to it to get anything out of it. Like turning it into the Quantity Theory of Money: In the long run, Y grows at a trend rate, V is stable. So in the long run an X% increase in the money supply leads to an X% increase in the general level of prices.

It does tell us something useful though, if we think about demand-side business cycles. They happen when nominal spending falls and nominal variables all have to adjust downwards. Nominal rigidities prevent this happening quickly and so output falls instead and we have a recession. What does the equation of exchange tell us?

MV = PY = nominal spending. It tells us any fall in nominal spending - and so any demand-side recession - is due either to a reduction in the growth of the money supply by the monetary authorities, or a failure to offset increases in the demand for money.

Applying that to the current downturn, unless you think it's real shocks rather than nominal, it directs your thinking to the right places. We saw a collapse in home prices, we saw a financial crisis, a long and deep downturn ensued. If you want to claim A caused B caused C, ask yourself how? How did that shit result in a downturn? V fell (the demand for money increased) because of uncertainty and blah blah blah. Why wasn't it offset by the monetary authority? If nominal spending fell (and it did, 9% below trend), it's because the Fed allowed it to happen. If the downturn is long and deep because the economy has to adjust to a lower level of nominal spending, it's not because of a housing bubble or a financial crisis, it's because of tight money.

 

[Baruch Menachem]

Dude, what is the point of all this? 90% of the folks on the board have no idea what you are talking about and the 10% who do think you are full of BS

 

[Norman]

MV = PY = nominal spending. It tells us any fall in nominal spending - and so any demand-side recession - is due either to a reduction in the growth of the money supply by the monetary authorities, or a failure to offset increases in the demand for money.

Well any demand side recession is due to people not spending enough money. It doesn't matter how you break those spending components. Of course these equations don't answer why recessions are formed. There are some good reasons why demand and NGDP should fall and not letting it do so could be catastrophic, so I would not necessarily call recession a "failure".

Applying that to the current downturn, unless you think it's real shocks rather than nominal, it directs your thinking to the right places. We saw a collapse in home prices, we saw a financial crisis, a long and deep downturn ensued. If you want to claim A caused B caused C, ask yourself how? How did that shit result in a downturn? V fell (the demand for money increased) because of uncertainty and blah blah blah. Why wasn't it offset by the monetary authority? If nominal spending fell (and it did, 9% below trend), it's because the Fed allowed it to happen. If the downturn is long and deep because the economy has to adjust to a lower level of nominal spending, it's not because of a housing bubble or a financial crisis, it's because of tight money.

Well I think it's pretty simple. People realize houses are way overpriced -> a lot of people quit their jobs realizing what they are producing isn't worth anything anymore.

So a bubble certainly isn't some monetary / NGDP problem that can be fixed via right kind of demand policy. There needs to be real shifts in employment. In this case less people need to be working to produce granite tops and houses, or selling mortgages etc.


If the FED really did the right thing they would have stopped the demand from increasing in the earlier 2000s before the bubble burst or started to form. That way people would have worked in more productive places to begin with.

 

[editec]

Ifritzme and DSGE...ya'll might want to check out


"Econ Point Forums - where economists discuss Economics"


There isn't a whole lot of activity there, now, but your input might be better recieved there where the audience's focus is on the nuts and bolts of economics.

Not that I in any way want you to stop posting here, of course.

 

[DSGE]

MV = PY = nominal spending. It tells us any fall in nominal spending - and so any demand-side recession - is due either to a reduction in the growth of the money supply by the monetary authorities, or a failure to offset increases in the demand for money.

Well any demand side recession is due to people not spending enough money. It doesn't matter how you break those spending components.

The point is that nominal spending is controlled entirely through monetary policy. Nominal spending is MV. The central bank controls M, so nominal spending is determined by the central bank.

Of course these equations don't answer why recessions are formed.

No, nominal rigidities does, for demand side recessions. An identity on its own doesn't tell you anything, but in the context of nominal rigidities it can tell you something.

There are some good reasons why demand and NGDP should fall and not letting it do so could be catastrophic, so I would not necessarily call recession a "failure".

...such as? That's ridiculous. There are good reasons why demand for certain goods and services should fall. There's no reason to prevent aggregate demand from falling. But there is a good reason to not let it fall.

Well I think it's pretty simple. People realize houses are way overpriced -> a lot of people quit their jobs realizing what they are producing isn't worth anything anymore.

And that's exactly the "recession" that should have happened. Resources were "misallocated", there should have been unemployment in the housing sector. That happened as home prices started declining in 2006. Residential construction employment tanked. It wasn't until 2008, when NGDP was allowed to collapse, that employment everywhere fell. Overpriced housing doesn't explain an economy-wide long and deep downturn. A collapse of nominal spending does.

So a bubble certainly isn't some monetary / NGDP problem that can be fixed via right kind of demand policy.

Well that's exactly the point. A "bubble" can't explain what we observe. What we observe is however entirely consistent with tight monetary policy.

There needs to be real shifts in employment. In this case less people need to be working to produce granite tops and houses, or selling mortgages etc.

Sure. So why is everybody else unemployed?

If the FED really did the right thing they would have stopped the demand from increasing in the earlier 2000s before the bubble burst or started to form. That way people would have worked in more productive places to begin with.

They kept aggregate demand growing smoothly. They can't control the relative demand for goods, and nor should they try.

 

[DSGE]

Ifritzme and DSGE...ya'll might want to check out


"Econ Point Forums - where economists discuss Economics"


There isn't a whole lot of activity there, now, but your input might be better recieved there where the audience's focus is on the nuts and bolts of economics.

Not that I in any way want you to stop posting here, of course.

Awesome. I've been trying to find something like that. Thanks.

 

[itfitzme]

Ifritzme and DSGE...ya'll might want to check out
"Econ Point Forums - where economists discuss Economics"
There isn't a whole lot of activity there, now, but your input might be better received there where the audience's focus is on the nuts and bolts of economics.
Not that I in any way want you to stop posting here, of course.

Excellent

 

[itfitzme]

MV=ΣPQ

The equation in this form actually tells us nothing about real world. It can never tell 1 dollar = 1 bread. The sides most always be the same.

Thus it can't be used to predict inflation. Since you actually need real goods to measure inflation. This equation happens entirely in the nominal world. Nominal GDP = Nominal GDP measured in 2 different ways. Velocity of money * money stock = all the money spent = NGDP and Average price * goods sold = all the money spent = NGDP. As you can see you can't tell if price of bread or prices in general increased using that, for that you do need to live in real world. NGDP =! real gdp.

I think the equation is a bit unnecessary when there is already much better way to understand price system - supply and demand.

Not only do you have to live in the real world, you have to live in all of it. Ignoring one part of the real world leads to conclusions that are not real.

If your looking for real GDP, all you do is

MV/CPI=Σpq/CPI=NGDP/CPI=R_GDP.

Σpq = p1q1+p2q2+p3q3 and those are real goods.

All those ps and qs are the equilibrium points of supply and demand.

M is the real money used in the real price of real goods purchased as this real money is used a real number of times. A

The equation, in this form, tells us some important things about what the real economy can do. It provides a framework within which supply and demand must function. It is an expression that accounts for all of the real world. And, it is really a real expression just as a sentence with words is a real expression. In fact, we can just as well put it into words and say that the real money that real people really use times the real number of times that they use it is identical to the real sum of the real product of real prices times the real quantity.

It is not so different than the physics equation of ΣF = Σma, where m is the total mass, "F" is the force exerted and "a" is acceleration. If we have a body made up of components with individual forces being applied, then "a" is the same across all of the "m"s so ΣF = aΣm.

This gives us that a=ΣF/Σm. And if we are constrained by the acceleration needs on some complex object, it puts a framework into how a change in one is accompanied by a change in the others. This is a real expression of real objects that really accelerate under the influence of real forces.

Farady, a real person that showed how real magnetism really affects real electricity did it with just words. But it's a pain in the a as a mathematical expression, like a picture, is worth a thousand words.

With MV=Σpq, one obvious thing that goes unstated when Mitt Romney says, "Bernanke has over inflated the currency", is that an increase in M can result in an increase in productivity just as well as resulting in an increase in prices. What Mitt is talking about when he implies that monetary inflation will lead to price inflation is the equation of exchange. And if our current problem is unemployment and a lack of output, then increasing the monetary base is necessary to allow output to increase. The problem with Romney's statement is it doesn't account for all of the real world and all the real effects that happen in the real economy.

Just as well, it is really true that if the money supply is not increased, it means an increase in prices must really be offset by a decrease in productivity.

When we look at what M is, exactly, it also says that an increase in the monetary base does not lead directly to either productivity or price increases simply because the M is not the monetary base. In order for an increase in the monetary base to result in a change to p and q, it has to get into the money flow, the wallet of the consumer that is paying for q at price p. If it never gets there, nothing happens.

This point is apparent when we look at the monetary base vs the CPI. I plotted the two with lag times from none to twelve months. All come up pretty much the same. I show this seven month one because it is the only one which has CPI increase with the large changes in the monetary base.

CPI is the change in prices on one side of the equation. The monetary base is the thing that gets M to increase. The relationship that the scatter plot shows is that CPI varies randomly with little "concern" for the change to the monetary base. This is really what really really happened.

If we are interested in real dollar or per capita, we can take MV=Σpq and normalize for population

MV/pop=Σpq/pop=NGDP/pop.

We can normalize for inflation with

MV/CPI = Σpq/CPI = NGDP/CPI = RGDP.

We can do both together, as in

MV/(CPI*pop)=Σpq/(CPI*pop)=NGDP/(CPI*pop)=RGDP per capita

which is, basically average standard of living. Better yet, we can divide by employment numbers

MV/(CPI*emp)=Σpq/(CPI*emp)=NGDP/(CPI*emp)=RGDP per worker,

which is basically productivity.

MV=Σpq also tells us something else as it defines the number of degrees of freedom for the "p"s and "q"s. The same equation is as functional in describing real Susan at the real store really shopping for the real family within a real limited budget. V is, in her case, 1. M is the real money in her purse. As she goes about deciding what she can really afford to buy, when she gets down to the last real price and the last real quantities of apple, she is really limited in how much she can really buy or how much she can really pay for that last item. Given a choice of generic or brand name, she can pick the one with the lower price for the sake of a larger quantity, or vis a vis.

However the prices and quantities work out as the forces of supply and demand find balance, they are constrained within the larger framework of MV=Σpq. It tells us that, all other things being equal, if the price of gasoline goes up, the price of bread must go down. That, or the quantity of DVD's sold must fall.

MV=Σpq is one of a number of real constraining relationships for the real economy. There are real manufacturing processes that determine the real minimum price for any given q. It should be obvious that electronic money has allowed V to increase considerably. It should be obvious that because the real guy that owns the real market down the street doesn't have to really walk from his store to the bank everyday that the real information about the money can get to the bank really quick.

We can divide the economy into smaller subsets. Real, geographic boundaries are workable as wealth tends to flock with wealth. If we consider that automobiles are sold across the US while Detroit is locally balanced in the rest of it's economy, we can represent it as

MV(Detroit)+MV(everyone else)=Σpq(Deroit)+Σpq(everyone else) (assuming V is the same complicated if it is different)

And, without doubt,
M(All of it)=M(Detroit)+M(everyone else)=Σpq(all of it)
MV(Detroit)=Σpq(Deroit)
MV(everyone else)=Σpq(everyone else)

If the amount of money flowing into Detroit is higher than that flowing out, then M(everyone else) must be reduced.

MV(Detroit)=Σpq(Detroit), within it's local boundaries. If M is increasing in Detroit then Σpq(Detroit) is increasing in terms of prices or local output.

MV(everyone else)=Σpq(everyone else) must balance so local prices and output, must fall.

We may just as well differentiate by real markets with MV(healthcare)=Σpq(healthcare) being one boundary. The same reasoning applies. It should be obvious that, like automobile manufacturing in Detroit, Oil production in Texas and Oklahoma, even healthcare providers tend to differentiate along geographic boundaries. In Fremont, California, doctors that work at the local hospitals and clinics do not live there, the live in Brentwood and other affluent neighborhoods. The prices of products, the Σpq(healthcare) are driven my the M of the doctors in Brentwood while the Σpq(everyone else) is driven by the local wealth within Fremont.

In terms of "predicting" prices and output, the change in Σpq is constrained by change in M.

It's not so different than considering the performance of an automobile engine. Yes, the carburetor is a significant part of the performance. When the accelerator is pressed, the potential demand for air and fuel is satisfied as the supply of air is increased. But it fits within the framework of the entire system. The flow of gases respond to the exhaust system just as much as they respond to the carburetor. If the exhaust system is restricted, no manner of pressing on the accelerator will get the car to go faster. We can look at the system by detailing every single element. And we can look at it from the perspective of the entire thing. If the car is sitting in a sealed garage that is filling with exhaust fumes with the oxygen depleting, seeing the big picture of the whole car within the garage tells us things that just tweaking the needle valve doesn't. If there is no air coming in, then all the accelerator pressing and needle valve turning isn't going to make the engine go faster.

The trick then, with MV=Σpq, because it absolutely describes the entire system, is in determining what exactly M and V is, as it changes from one time to the next. If they don't change, the forces of supply and demand that have micro effects in that prices and quantities, Σpq, are constrained by the macro effect of M and V.
This is helpful because there are 160+ million workers, 310+ million people, and 7+ million businesses.

It is in this issue that I find the trouble. The accumulation of consumer credit affect M more directly than the monetary base does. It is, in fact, a direct measure of one portion of M. After all, if it a measure of money used in purchases. As output or prices increase, that increase can be entirely absorbed by an increase in consumer credit that increases M.

And if we look at two particularly big recessions, the recent 2007 recession and the recession of we find them marked by consumer credit not increasing. In the recession of 1990, the major restriction was on non-revolving credit. In the 2007 recession, it was both revolving and non-revolving credit that froze up.

Because MV=Σpq puts an absolute constraint on the macro economy, if M decreases, then Σpq must decrease. And with consumer credit being a source of M, a decrease in consumer credit means output or prices must decline. If prices just won't go down for either macro or micro reasons, then output must fall.

The flip side is that if M is dependent entirely on consumer credit to account for increases in pricing and output, then we have a system which requires that the Federal government or the middle class consumer accumulate debt.

The equation of exchange isn't better than supply and demand constraints. Nor are functions that express S-D better than EOE. Individual markets of p and q are constrained by both the forces of supply and demand along with the macro economy in which they function. They are complimentary. Supply and demand fits within the framework of the EOE. The monetary base, M1, M2 and other measures of money fit within its framework. The business economic constraints on production efficiency and output fits within its framework. The change in consumer credit fits within it's framework.

All of these micro effects fit within the frame work of the EOE. As well, taken alone, it presents some information about how the macro economy can function in terms of price inflation or productivity increases going hand in hand with changes in the money supply.

 

[itfitzme]

Dude, what is the point of all this? 90% of the folks on the board have no idea what you are talking about and the 10% who do think you are full of BS

If I lived my life based on what 90% of people understand or the other 10% that think it's bs, I wouldn't have learned how to change the oil in my car or design a heater control.

Whether we realize it or not, 99.9% of what we know we have learned from someone else.

Had I approached things from "I don't understand it and it is bs", I would have missed a thousand opportunities to understand things that I wouldn't have otherwise. I've gotten as much out of trying to prove that "Bernanke over inflated the currency" and finding it is wrong as I do from some one else that says, "I'm a pure math guy and a better approach is Friedmans...." In a few short sentences, he opened up an avenue that I might have not "reinvented" in a year of consideration or might have otherwise passed over in reading Wikipedia. Sure, I can do the math, he's got better insight. Physics and engineering would be lost without the pure math guys. The pure sciences produces a product that it sells to the applied sciences that produces a product that then sells to the sales and marketing guys who then have a product they can sell to the consumer.

If everyone measured by the 90% and 10% we wouldn't have GPS systems because, when you read Einsteins 1904 paper on Special Relativity, it is not easily understood and might just be bs. I mean, common, clocks slow down in orbit because gravity warps space-time? If that doesn't sound like a bunch of bs, then nothing does. Thankfully, someone smarter than me understood it and explained it more simply. That eventually led to companies like Trimble Navigation and Magellan.

And seeing as I have to pay money for the opportunity to learn from the likes of John T. Harvey or Krugman, who are way beyond my level, I am nothing less than grateful that Congress, including Al Gore, were instrumental in organizing the internet when businesses recognized they had an opportunity to open up new markets through online sales. I get the opportunity to read what other people have written, whether I understand it or not.

I am especially grateful because I live with one of those 90% who is such a dumb shit that he can't even read the want ads to get a new job and has to dig through the garbage behind Food Max to get a piece of old meat so he ends up vomiting all over the bathroom floor. After all, in his opinion, it's all just bs anyways, so why bother learning.

 

[itfitzme]

Yeah so I don't agree with the part about M being "the stock of money in flow".

So how is this consistent with the $1.6 trillion in free reserves but no hyperinflation? Because of the "liquidity trap". Liquidity trap theory tells us that when the interest rate on bonds becomes zero, the demand for money becomes perfectly elastic. Any increase in the monetary base will be offset by an equal increase in the demand for money (decrease in velocity). So dM = -dV.

Of course there are a myriad of problems with the liquidity trap stuff, but it gets to the point.

This is very interesting.

Well, if you don't see it as the M that is used in the exchange for the P then it's a bit of a sticky wicket. You can always go with Keynes who "argued that a certain portion of the money supply will not be used for transactions, but instead it will be held for the convenience and security of having cash on hand" My sense of it is that either Hume wasn't clear enough and Mills missed it. So Keynes had to add the k because, by that time, it had kind of mutated.

So we could go with Keynes and use (M/k)V=PQ, but I don't see the point. We can use Mflow to mean (M/k) and then have Mflow*V=PQ. That is why I keep using M-flow.

But, by definition of MV=PQ, M refers to the sum of all the monies used in the exchanges as P without counting them twice. M only represents something else if 1/k is introduced. But then, were just adding in another unknown on top of the immeasurable M and V.

Regardless of Hume, Mills and Keynes, by the same fundamental process that produces F=Ma, V=RI, and a host of other equalities, MV=PQ only if M refers to the sum of all the money used for P without counting any individual piece twice.

And, while dM = -dV is one thing that can happen, it isn't the only thing that can happen. Obviously, when the statement is made that "the govt' printing money causes inflation" isn't the only thing that can happen. Clearly, it can also allow for the increase in productivity. Really, where you point out that dM=-dV, Romney claims that dM = QdP. And an alternative is that dM = PdQ.

MV = PQ says that if M changes, any one or combination of V, P and Q can change to account for it.

But what holding to M is the money that is used in the purchases of Q is that we can then go looking for exactly what it consists of.

Maybe it's a cultural difference between business and engineering. One instructor of management, in discussing power and authority, pointed out that engineers follow an authority that resides outside the political structure of the company. This usually frustrates non engineer managers. They just don't understand why the engineer keeps telling him "no". No one tells a manager "no". He's the authority of all that is. And yet engineers do, there is simply the authority of physics that doesn't bend to the will of the manager. There are authorities that have to do with market standards. There are legal authorities that have to do with product safety.

I have noticed that the major market for economists products is business and government. Business and government tends to not see physics as a final authority. They tend to have to define every word at the beginning of the contract or law because the other guy will decide he thought it meant something else.

It's different in engineering and physics where every lecturer is required to prove everything presented with no resting on his laurels. If he can't prove it by deduction, he can always prove it in lab. F=Ma doesn't care if he has a bachelors or PhD. And neither do the students.

In working out how fast some rocket will accelerate, it does little good to include the fuel that might have been use but is sitting in a storage tank or consider the mass to include a couple of extra mice that might have been added to the payload but wouldn't fit in the nosecone. The M refers to the sum of all the little mice in the payload. The F means the output of the engine by the fuel in engine. So a=F/M is the result of the Fs and the Ms that are used to create the "a". If "a" referred to anything else but what is sitting on the other side of that equal sign, the equation is meaningless.

So I'm left to scratch my head when I run into "I don't agree with the part about M being "the stock of money in flow"." As I don't know how else we can take M in MV=PQ if not by that is means the money actually used to pay the price p for the quantity q of products during the exchanges without counting anything twice.

After all, that's how Copernicus did it, that's how Volta did it, that's how Einstein did it. And if that's not how Humes did it, I'm not surprised that economics is often referred to the dismal science.

It's just darned interesting.

 

[DSGE]

This "stock in the flow" way of thinking about it is convoluted. I think about as M being a stock of money, any money stock you like. V is the average number of times a dollar from that money stock gets used in an exchange. Simple. If some of the money stock gets held, that just lowers V.

I figure you're thinking about it like... there's a stock of money. If a dollar gets used in a transaction, tag it, and trace how many transactions it goes through in total. M is the stock of all tagged dollars. V is (1/M)Σn, where n is the number of times each dollar gets used in a transaction.

It's easier to just think about M as the total money stock and have n=0 for money that isn't spent. There's no reason at all to condition on "the money gets into the flow".


As for the other stuff, physics has had a few hundred years to get all its definitions worked out so that they do interesting things. Econ is reasonably young. You could probably cut it some slack.

 

[itfitzme]

This "stock in the flow" way of thinking about it is convoluted. I think about as M being a stock of money, any money stock you like. V is the average number of times a dollar from that money stock gets used in an exchange. Simple. If some of the money stock gets held, that just lowers V.

I figure you're thinking about it like... there's a stock of money. If a dollar gets used in a transaction, tag it, and trace how many transactions it goes through in total. M is the stock of all tagged dollars. V is (1/M)Σn, where n is the number of times each dollar gets used in a transaction.

It's easier to just think about M as the total money stock and have n=0 for money that isn't spent. There's no reason at all to condition on "the money gets into the flow".


As for the other stuff, physics has had a few hundred years to get all its definitions worked out so that they do interesting things. Econ is reasonably young. You could probably cut it some slack.

Obviously, the field of economics didn't really begin until after the Great Depression with became the motivator for collecting much more economic data than was previously available. And economics has the difficulty of needing to work with what it has, the data that is collected by govt agencies and the natural experiments as they occur.

--

I tracked back to Hume and found his original essay at

http://93beast.fea.st/files/section1/hume/extras/Essays, Moral, Political, and Literary.pdf

An earlier post said that supply and demand are the thing. A quick read of Hume seems to state that, in fact, all those prices are dependent on the money supply. I am actually of the macro econ kind but it's examination let me to consider how they connect. And, in fact, that PQ in MV=PQ are the equilibrium points in the markets.

What may be missed is that

MV=PQ = PQ(Food) + PQ(Energy) + PQ(HealthCard)+PQ(whatever)

When the price for one product increases, given a constant money supply, it necessitates that others decrease. Micro markets do not act in isolation. If M is fixed and PQ(Energy) goes up, then everything else must go down.

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

I was reading another blog where someone referred to MV=PQ as a tautology

With V defined as GDP/M, it is as GDP is PQ

The result, is that of course v decreases with an increase in M, V provides no more information

Simply,

MV=PQ = GDP

Substituting GDP/M for V, or more directly, PQ/M we get

M*(PQ/M)=PQ which, when all is canceled just yields

1=1

That V decreases proportionally to an increase in the money supply is by definition.

No manner of definition for M will solve this problem as long as V isn't independently determined or constant.

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

But, at least now I have a better understanding of what is up with the V=GDP/M1 etc. And it does require V to include more than just the frequency of exchange in the purchases. It becomes a multiplier that includes both the frequency of exchange and forces that affect the demand for money.

On the other hand, if we stick with

M_flow= Spendable_Income + ΔCredit - ΔSavings then V is just the frequency of exchange.

If I might note, monies that come from investment monies, either borrowed, the result of an IPO, or however through the money multiplier, becomes a factor that affects both sides of MV=PQ. As the source of capital equipment purchases, it increases the potential for output of Q. At the same time, it becomes an increase in Income which increases demand potential.

Mbase -> M_investment -> M_flow and it gets there through numerous different paths.

That alone seems to suggest some utility to the approach because it delineates out that liquidity trap from other factors.

At this point, I'm probably going to see what V is when M is restricted to M_flow. I don't know if, during Keynes times, there was enough data. But now we have all of it, income, credit, savings, and taxes.

 

[itfitzme]

This "stock in the flow" way of thinking about it is convoluted. I think about as M being a stock of money, any money stock you like. V is the average number of times a dollar from that money stock gets used in an exchange. Simple. If some of the money stock gets held, that just lowers V.

I figure you're thinking about it like... there's a stock of money. If a dollar gets used in a transaction, tag it, and trace how many transactions it goes through in total. M is the stock of all tagged dollars. V is (1/M)Σn, where n is the number of times each dollar gets used in a transaction.

It's easier to just think about M as the total money stock and have n=0 for money that isn't spent. There's no reason at all to condition on "the money gets into the flow".

Going back to Hume who is reported to have first considered this function of exchange, he says,

"It is also evident, that the prices do not so much depend on the absolute quantity of commodities and that of money, which are in a nation, as on that of the commodities, which come or may come to market, and of the money which circulates. If the coin be locked up in chests, it is the same thing with regard to prices, as if it were annihilated; if the commodities be hoarded in emagazines and granaries, a like effect follows. As the money and commodities, in these cases, never meet, they cannot affect each other. Were we, at any time, to form conjectures concerning the price of provisions, the corn, which the farmer must reserve ffor seed and for the maintenance of himself and family, ought never to enter into the estimation."

 

[Norman]

The point is that nominal spending is controlled entirely through monetary policy. Nominal spending is MV. The central bank controls M, so nominal spending is determined by the central bank.

Well Central bank can create and spend money so they can influence it to the point of controlling it, yes. No disagreements here.

...such as? That's ridiculous. There are good reasons why demand for certain goods and services should fall. There's no reason to prevent aggregate demand from falling. But there is a good reason to not let it fall.

Sorry but I don't think there is a constant of "NGDP". The number at which it should always be for maximal wellbeing given. How do you know you have the right constantat X anyway? Maybe the best economic outcomes come if it's the money that is constant. Or maybe not. I am saying that you are assuming.

If the FED really did the right thing they would have stopped the demand from increasing in the earlier 2000s before the bubble burst or started to form. That way people would have worked in more productive places to begin with.

They kept aggregate demand growing smoothly. They can't control the relative demand for goods, and nor should they try.
[/QUOTE]


So with this in mind are you saying NGDP should be constant or only increase.

Either way, we are just blocking the discussion here. Let's move it somewhere else.

 

[Norman]

Not only do you have to live in the real world, you have to live in all of it. Ignoring one part of the real world leads to conclusions that are not real.

If your looking for real GDP, all you do is

MV/CPI=Σpq/CPI=NGDP/CPI=R_GDP.

Σpq = p1q1+p2q2+p3q3 and those are real goods.

Uhm. No you don't. When you insert the CPI there you start living in real world, but that completely changes the equation and renders it meaningless.

CPI can change while M and V are both stable so you can not at that point predict anything anymore.

You can not use the equation to predict inflation. But it's useful if you want to know how sizeful an average transaction is (rather than how much an average transaction buys). How much is it BTW?

Anyway I will check the links...

When the price for one product increases, given a constant money supply, it necessitates that others decrease. Micro markets do not act in isolation. If M is fixed and PQ(Energy) goes up, then everything else must go down.

Incorrect we live in a world where wealth is not a constant. So even if all the monetary factors are constant, prices may not be or you are violating law of supply and demand. This means all prices can fluctuate even in aggregate to up or down in whatever monetary situations. I just think it is way better to think in terms of supply and demand than this equation. NGDP is not real GDP.

Also this is incorrect because according to the equation it is also possible that when price of a product halves you buy 2x more of it. If you "halve" the price of potatos from 2 per 2 dollar to 53243 per 1 dollar nothing happens if people buy 2 times more of those 1 dollar transactions. See what I did there? Nominal tinkering gets us nowhere when measuring REAL inflation.