Global Economic
Trend Analysis

Recent Posts

Tuesday, January 09, 2007 12:06 AM

Money Supply and Recessions

Mish Moved to MishTalk.Com Click to Visit.

This post is an attempt to construct an active chart of money supply data that is useful in assessing the likely direction of future economic activity. This is an ongoing effort that relies heavily on the work of Austrian economist Frank Shostak who has previously written on this subject on the Ludwig von Mises Institute and other places.

Before we can construct charts of money supply we first must agree what it is we want to measure. Answers.com provides many definitions definition of money but none of them are useful for our purposes. Wikipedia's disscussion of Money Supply includes M0, M1, M2, and M3. There is also MZM (Money with Zero Maturity). So which one of those makes the most sense to track?

Please stick around because the answer might surprise you, with lots of neat charts to prove it (click on any chart for a better look at it) . But first let's tackle the methodology.

In The Mystery of the Money Supply Definition Shostak wrote:

The crux in identifying what must be included in the money supply definition is to adhere to the distinction between a claim transaction and a credit transaction.

According to popular thinking, the inclusion of savings deposits into the money supply definition is justified on the grounds that money deposited in saving accounts can always be withdrawn on demand. But the same logic should also be applied to money placed with an MMMF [Money Market Mutual Fund]. The nub, however, is that savings deposits do not confer an unlimited claim. The bank could always insist on a waiting period of thirty days during which the deposited money could not be withdrawn.

Savings deposits should therefore be considered credit transactions with depositors relinquishing ownership for at least thirty days. This fact is not altered just because the depositor could withdraw his money on demand. When the bank accommodates this demand, it sells other assets for cash. Buyers of assets part with their cash, which in turn is transferred to the holder of the savings deposit. The same logic is applicable to fixed-term deposits like CDs, which are credit transactions.

Though traveler’s checks are considered an integral part of the money supply, they should not be. Traveler’s checks are receipts for investment in the companies that issue them. As such, they result from a credit transaction, and therefore are not part of the money definition. Cashing a traveler’s check means that AMEX or VISA will transfer money from their deposits to the holder of the check, which will not change the amount of money in the economy.

Mainstream thinking currently excludes from the money supply government deposits held in banks and the central bank. Consequently, if the government taxes people by one billion dollars, money is transferred from their deposits to the government’s deposit. This is viewed just as if the money supply fell by one billion dollars. In reality, however, the money is now available for government expenditure, meaning that money held in government deposits should be part of the definition of money.

Incorporating all the above arguments, the money supply is defined as follows: Cash + demand deposits with commercial banks and thrift institutions + government deposits with banks and the central bank.

This definition shows clearly that any expansion in money supply results solely from central bank injections of cash and commercial banks’ fractional reserve banking.
In Making Sense of Money Supply Data Shostak further explains the flaws of M2, M3, and MZM as monetary measures:
Consider M2. This definition includes money-market deposit accounts. However, investing in a money market fund is in fact investment in various money market instruments. The quantity of money is not altered as a result of this investment; only the ownership of money has temporarily changed.

If Joe invests $1,000 with a money market fund, the overall amount of money in the economy will not change as a result of this transaction. Money will move from Joe's demand deposit account to a money market demand deposit account with a bank. To incorporate the $1,000 invested with the money market fund into the definition of money plus the original $1,000 would therefore amount to double counting.

The problem of double counting is also not resolved by the money of zero maturity definition of money (MZM)—a relatively recent money supply definition. The essence of MZM is that it encompasses financial assets with zero maturity. Assets included in MZM are redeemable at par on demand. In short, MZM includes all types of financial instruments that can be easily converted into money without penalty or risk of capital loss. This is precisely what is wrong with this definition, since it doesn't identify money but rather various assets that can be easily converted into money. In short, it doesn't tell us what money actually is and where money is located, which is what a definition of money is supposed to do.
From the above we can see that M2 double counts some aspects of money and since M3 incorporates M2, it is equally flawed right off the bat. M2 also counts savings accounts which are really credit transactions and MZM is hopelessly flawed as a money definition because it attempts to count things that can easily be converted to money as opposed to money itself.

M1 is flawed by the inclusion of travelers checks and the exclusion of demand deposits with commercial banks and thrift institutions + government deposits with banks and the central bank. M1 also has another huge problem and that problem is sweeps.

Returning to The Mystery of the Money Supply Definition we see:
Since January 1994, banks and other depository financial institutions have initiated sweep programs to lower statutory reserve requirements on demand deposits. In a sweep program, banks “sweep” funds from demand deposits into money market deposit accounts (MMDA), personal savings deposits under the Federal Reserve’s Regulation D, that have a zero statutory reserve requirement ratio. By means of a sweep, banks reduce the required reserves they hold against demand deposits.
Essentially banks are taking money that should be available immediately on demand (and expected to be demanded sooner rather than later by the very nature of checking accounts) and loaning it out anyway. That money is missing from M1 published demand deposits numbers. Unfortunately sweeps data only comes out once a month (at least from the free sources at my disposal) but once a month is probably often enough for most purposes.

The definition I am now working with is Cash + demand deposits with commercial banks and thrift institutions + government deposits with banks and the central bank + sweeps + other checkable deposits. I added other checkable deposits because with their inclusion I could better match previous publicly posted Money AMS charts by Shostak.

I received charting help from a couple of people that I now need to thank (originally "Yaniv" followed by "Bart" at NowAndFutures). Since we are not sure if this exactly follows MoneyAMS (in fact we are sure that we are slightly off MoneyAMS from 2003 on), we are calling our indicator M' and pronouncing it M Prime.

Let's see how M' does compared to M2 as a predictor of economic activity.

M Prime

Leaving the current status as unknown, 6 of the last 6 recessions were marked by a major dip in M Prime. Note how the indicator clearly led the recession. Also note that 6 of 8 sustained dips below an annual growth rate of 5% in M' led to a recession. On that basis we have 2 potential false signals (1985 and 1995).


Unlike M', the direction of M2 does not seem to give clear economic signals. Note that M2 was rising into the double dip recession of 1982 and rising into the 1991 recession as well. Also note that the single largest dip in M2 was in 1993 while M' was soaring. The years between 1992 and 1995 are all problematic. Finally note that unlike M' where a dip below 5% annual growth was a huge warning sign, the dotted line above shows no such significance. M' seems to be far superior to M2 as a leading indicator.

M Prime CPI Adjusted

On a CPI adjusted basis we see that there has been a recession on 6 of 7 sustained dips below the zero line of year over year growth in M'. The 1985 excursion below 0% was extremely brief in stark contrast to all of the labeled recessions and thus can be discounted. 1995 is still a miss but nowhere near as pronounced as compared to M' unadjusted. 1995 also happens to correspond to the start of a huge ramp-up in sweeps. Perhaps that is significant and perhaps not. Nonetheless, M' CPI adjusted gives a cleaner signal, arguably calling for 7 recessions of which 6 happened.

The above chart clearly shows M' CPI adjusted to be a strong leading economic indicator.

M2 CPI Adjusted

M2 CPI adjusted is certainly an improvement over M2 CPI unadjusted. Note however, that the 1982 and 2001 signals are not as strong as the corresponding M' CPI adjusted signals. The M2 adjusted signal for 2001 was particularly weak. More problematic for M2 adjusted vs. M' adjusted is the mass of jello between 1988 and 1996. M' adjusted was clearly giving an "all clear" zero cross signal by 1992 while M2 adjusted gyrated for years and did not really give an "all clear" until 1998. Furthermore M2 adjusted actually dipped back below the zero line in 1997 while M' adjusted was soaring upward. Both M2 and M' missed around the 1996 timeframe but even here M2 did worse both in terms of an actual low and the jello that preceded it.

Since everyone seems to be focused on M3 here are the two corresponding M3 charts for the above series.


M3 CPI Adjusted

As leading economic indicators, the above charts on M3 fared worse than M2 and far worse than M Prime. M3 CPI adjusted failed as a leading indicator in 3 out of 6 recession but did act as a lagging indicator in the 1973 to 1975 recession. Perhaps this lends credence to the Fed's statement that they see little use for M3. Still it is easy enough to post the figures since they have the data.

One thing you can say about M3 is that it that for all this credit expansion, we sure do not have a lot of GDP growth to show for it. The real economy is not benefiting.

The CPI Adjustments

One of the problems in constructing the CPI adjusted charts is believing the CPI number itself. Clearly there is a lot of disagreement with the CPI numbers. Yet without the adjustment, using M2 as any kind of leading indicator is nearly useless. In contrast, M' is a very good leading indicator in and of itself, but it seems to be a better one when adjusted. Yaniv, Bart, and I tried various combinations of including or excluding other monetary figures but none of them performed as well as what Shostak proposed as theoretically ideal.

In one test we tried using alternative estimates of CPI based on John Williams' Shadow Statistics. Those tests seemed to make some things worse and some things better. Here are a couple Shadow Statistics charts to consider.

Shadow Statistics

The first chart shows an approximate average of the Official CPI and the Shadow to be running at approximately 5%. The second chart shows an approximate 5% trendline using Pre-Clinton measures of CPI. (The dashed trendlines were added by me). Meanwhile the long term annual rate of change on M' and M2 since 1969 are 6.5% and 6.9% respectively. All of this calls into question what the Fed is really doing.

Regardless of what the Fed is or is not doing (purposely or otherwise), it is impossible for them to succeed with price targeting. This idea was discussed in Is the Inflation Monster Tamed? when the following questions were asked:
  • Why should inflation be targeted at 2% and not 1% or 3%?
  • Why should any inflation be targeted at all?
  • Even if it was for some reason smart to target prices, can prices really be measured it accurately?
  • What do central banks do to overcome lag effects of monetary tightening and loosening?
  • Is this just blind faith "we know neutral when we see it"?
Who benefits from inflation?
Inflation benefits those that receive money first: the government and banks. The former is via automatic tax increases not indexed to inflation (especially property taxes), the latter simply because banks are first in line to receive money from the FED at rates no one else sees. By the time lending standards drop so that the masses have access to credit, the boom is well underway. By the time credit is granted to anyone that can fog a mirror, the boom is nearly over. Those buying assets late in the game will eventually be crushed by those selling assets that got in early. Simply put, inflation eventually becomes a moral hazard.
Proof that the Fed is not succeeding can be found in the endless series of bubbles that keep getting larger and larger, as well as the ponzi like scheme it is taking to keep things afloat.

As for leading indicators, the charts speak for themselves. Given that our starting point is the best Austrian approximation of money under the current fiat system that we could find, it should not be surprising that those monetary indicators give off far superior leading economic signals than other methods.

I wish to thank Frank Shostak for his enormous contribution to these efforts. I also wish to thank "Yaniv" who started this charting exercise with me a year ago or so, "Bart" who picked up when Yaniv ran out of time, John Williams at Shadow statistics, Scott Reamer on Minyanville, Paul Kasriel who offered some comments that I will share later in a post on leading indicators, and anyone and everyone that contributed to this effort.

Mike Shedlock / Mish

Last 10 Posts

Copyright 2009 Mike Shedlock. All Rights Reserved.
View My Stats