U.S. M1 VELOCITY INCREASED! BUT WHY? (Originally from 2015)

(I believe that most of the blog entry below is from August 19, 2015 with some information added August 28, 2015.)



According to recent estimates on the Saint Louis Federal Reserve web page and elsewhere, the velocity of the M1 money supply in the U.S. increased in the second quarter of 2015.  This was the first increase in years!

For those not familiar, velocity can be calculated by dividing nominal Gross Domestic Product (or nominal GDP) by the money supply.  Gross domestic product is the main measure of the production of goods and services in a country in a given time period.  Nominal GDP values each good or service produced at however much that item cost when it was produced, so that nominal GDP in the U.S. is measured in dollars that have varying buying power.  Real GDP values each good or service at its base-year price, so that U.S. real GDP is measured in dollars that have constant purchasing power.  (Real GDP is better than nominal GDP for comparing production between two time periods because changes in the price level have no impact on real GDP (as long as the base year does not change).)  Because nominal GDP equals real GDP (often abbreviated as Y) multiplied by the implicit deflator divided by 100 (representing the price level, P), velocity can be calculated as (Y*P)/M, where M represents the money supply.

Because the velocity of money in the U.S. shows how many times per year each U.S. dollar trades on average in nominal GDP transactions, increases in money velocity are often associated with economic growth, with dollars trading more frequently when velocity rises.  However, although both real GDP and the M1 money supply increased in the second quarter of 2015, the increase in the implicit deflator in that quarter would have caused the velocity of M1 to increase if real GDP and M1 would have remained unchanged.  Thus, an increase in the velocity of money does not guarantee that economic growth has occurred.

Moreover, while the increase in nominal GDP in the second quarter of 2015 contributed to the relatively small increase in M1 velocity in that quarter, another factor was a slowdown in the growth rate of the M1 money supply.  When both the M1 money supply and nominal GDP increase, the velocity of M1 increases if the percentage growth in nominal GDP exceeds the percentage growth in M1, and this is what happened in the second quarter of 2015 in the U.S.  If both M1 and nominal GDP increase by the exact same percentage, then velocity remains unchanged.  If both M1 and nominal GDP increase but the money supply grows at a faster rate than nominal GDP, then velocity falls.

In many other recent quarters in the U.S., nominal GDP has increased, but it has not increased nearly as rapidly as the M1 money supply.  Thus, the velocity of M1 has continued its decline during most quarters of the lackluster recovery, although this decline in M1 velocity started during the Great Recession.

The U.S. had been in a period of rapid money supply growth, with probable goals being increasing the growth rate of real GDP, increasing employment, and enhancing the solvency of the financial system.  Periods of rapid money supply growth in the U.S. have often been accompanied by decreases in the velocity of money, even during times when real GDP was increasing.  This is one of the themes in my book, It’s Velocity, Stupid!  Is the Velocity of Money the Forgotten Variable of Macroeconomics?  My book also points out that the velocity of the M1 money supply and the velocity of the M2 money supply sometimes move in opposite directions within the same quarter.  This happened in the second quarter of 2015 in the U.S. (although data from that quarter are not discussed in the book), with M1 velocity increasing as noted above but M2 velocity continuing to decrease.  (ADDED AUGUST 28, 2015:  Please see NOTE in parentheses below.)  Because nominal GDP increased in the second quarter of 2015, the fact that M1 velocity increased while M2 velocity decreased must have been due to M2 growing at a faster rate than M1 in the second quarter.  This illustrates further the potential importance of the money supply growth rate in determining the velocity of money, at least in the short term.  (In many recent quarters, M1 had grown more rapidly than M2, resulting in M1 velocity falling more rapidly than M2 velocity.)

Regarding the growth rates of M1 and M2 in the second quarter, did savings accounts, certificates of deposit, and/or other assets part of M2 but not part of M1 grow more rapidly than assets part of both M1 and M2 due to at least some interest rates rising in the second quarter of 2015 (with non-M1 parts of M2 paying higher expected returns than M1 components)?  If so, then what will happen in the third quarter of 2015 and beyond now that at least some interest rates have been decreasing?  To the extent that at least some interest rates have been decreasing, how long will that last?

Has the velocity of M1 started a period when it will increase most, if not all, quarters?  We will need to wait to find out for certain.  However, future money supply growth rates could be an important (and potentially overlooked) determinant.

(ADDED AUGUST 28, 2015:  NOTE:  After the initial posting of this blog entry on August 19, recent data revisions to nominal GDP, real GDP, and perhaps the implicit deflator have led to the velocity of M2 also increasing in the second quarter of 2015, rather than decreasing.  While economic growth stronger than originally reported certainly contributed to the reversal in the directional movement of the velocity of M2 from decreasing to increasing, the slower growth in the M2 money supply in the second quarter allowed the velocity of M2 to rise.  In fact, had the M2 money supply grown as rapidly in the second quarter of 2015 as it grew in the first quarter, then the velocity of M2 still would have decreased despite the upward revisions to nominal GDP and real GDP.)