The most common way money velocity is as the ratio of money supply to GDP.
In the monetarist equation MV=PQ, MV is generally understood to be equal to nominal GDP. With Q equal to the quantity of goods and services produced to generate GDP and P equal to the general price level, mathematically, inflation is when an increase in MV disproportionate to the increase in Q results in a disproportionate rise in P (e.g., if MV increases 20% but Q only increases 10%, the resulting final price level P is (1.2MV)/(1.1Q), which results in an inflation rate of 9%)
If GDP is growing rapidly and the money supply is held constant, one would see a dramatic increase on the money velocity.
Conversely, if GDP is growing rapidly and velocity is constant, there would have to be a dramatic increase in the money supply.
However, since the rate of GDP growth is not intrinsically tied to the money supply, one would not expect every dramatic increase in money supply to be matched by a proportionate drop in money velocity, nor every dramatic shift in money velocity to be matched by a proportionate shift in money supply.
Looking at the period 1962-1976, while from 1968 on we see increasing divergence between M1 money supply and M1 velocity changes, we do not see the level of symmetry that appears later.
From about the 3rd quarter of 1981 onward, and especially after the 2008 financial crisis, money supply and money velocity changes become highly symmetric.
One noteworthy aspect of this symmetry is that, where there are phases where the change in money velocity rises above the change in money supply (when the curves cross, in other words), consumer price inflation crosses both curves near the intersection. It is that particular behavior that serves to establish that the symmetry--and the resulting offsetting inflationary impacts of changes in money supply and money velocity--are what act as constraints upon inflation. If money supply growth is pushing inflation up, money velocity deceleration is pulling it down, and vice versa.
The symmetry is not at all a necessary adjunct to money supply behavior. In the mid 1960s there are periods where money supply changes and money velocity changes are both increasing simultaneously. We do not see such periods after 3Q 1981.
Velocity is fundamentally the number of times money changes hands. Which is why it's commonly expressed as the ratio of GDP to money supply--by definition, that ratio is a measure of how often money changes hands.
While we don't measure velocity directly, it is still an important concept, because that frequency of money changing hands is going to directly impact GDP level and inflation.
As the US experience with inflation post-Volcker demonstrates, velocity is quite impactful on inflation, and can at times be even more impactful than the level of the money supply itself.
How does one measure Velocity?
The most common way money velocity is as the ratio of money supply to GDP.
In the monetarist equation MV=PQ, MV is generally understood to be equal to nominal GDP. With Q equal to the quantity of goods and services produced to generate GDP and P equal to the general price level, mathematically, inflation is when an increase in MV disproportionate to the increase in Q results in a disproportionate rise in P (e.g., if MV increases 20% but Q only increases 10%, the resulting final price level P is (1.2MV)/(1.1Q), which results in an inflation rate of 9%)
If GDP is growing rapidly and the money supply is held constant, one would see a dramatic increase on the money velocity.
Conversely, if GDP is growing rapidly and velocity is constant, there would have to be a dramatic increase in the money supply.
However, since the rate of GDP growth is not intrinsically tied to the money supply, one would not expect every dramatic increase in money supply to be matched by a proportionate drop in money velocity, nor every dramatic shift in money velocity to be matched by a proportionate shift in money supply.
Looking at the period 1962-1976, while from 1968 on we see increasing divergence between M1 money supply and M1 velocity changes, we do not see the level of symmetry that appears later.
https://fred.stlouisfed.org/graph/?g=Vlpb
From about the 3rd quarter of 1981 onward, and especially after the 2008 financial crisis, money supply and money velocity changes become highly symmetric.
https://fred.stlouisfed.org/graph/?g=Vlpu
One noteworthy aspect of this symmetry is that, where there are phases where the change in money velocity rises above the change in money supply (when the curves cross, in other words), consumer price inflation crosses both curves near the intersection. It is that particular behavior that serves to establish that the symmetry--and the resulting offsetting inflationary impacts of changes in money supply and money velocity--are what act as constraints upon inflation. If money supply growth is pushing inflation up, money velocity deceleration is pulling it down, and vice versa.
The symmetry is not at all a necessary adjunct to money supply behavior. In the mid 1960s there are periods where money supply changes and money velocity changes are both increasing simultaneously. We do not see such periods after 3Q 1981.
That's some serious had waving rather than an answer to a straight forward question.
It seems to me that there is no way to actually *measure* V; that we can only calculate it by taking P times Q and dividing by M.
Velocity is fundamentally the number of times money changes hands. Which is why it's commonly expressed as the ratio of GDP to money supply--by definition, that ratio is a measure of how often money changes hands.
While we don't measure velocity directly, it is still an important concept, because that frequency of money changing hands is going to directly impact GDP level and inflation.
As the US experience with inflation post-Volcker demonstrates, velocity is quite impactful on inflation, and can at times be even more impactful than the level of the money supply itself.