|
There's a lot of confusion about money and about
what does and does not form part of the money supply. Our goal in this short
discussion is to reduce the confusion.
We were prompted to revisit this issue in today's report by the first few
paragraphs of John Hussman's 15th April missive. Although there aren't any glaring errors in Hussman's
money-supply comments, they add to the confusion by failing to properly
distinguish between bank reserves, electronic money ("bits and
bytes") in bank checking and savings accounts, and physical currency
(notes and coins) in circulation. For the purposes of this discussion we'll
refer to electronic money in bank accounts as deposit currency.
The first point to be understood is that in the US economy only about 10% of
the economy-wide money supply, as determined by the True Money Supply (TMS)
calculation, is in the form of physical notes and coins. The rest is deposit
currency. Consequently, if percentages remained the same then there would
usually be about 9 dollars of deposit currency added to the money supply for
every new dollar of physical currency. For example, with the amount of
physical currency in circulation having increased by $320B dollars since the
Fed commenced its "quantitative easing" in September of 2008, it
would be normal for the amount of deposit currency to have increased by about
$2.9T over the same period to give a total TMS gain of about $3.2T. We
calculate that TMS actually increased by $3.7T over the period in question,
which is in the right ballpark and close to what we would expect considering
that the public's need for physical currency would not have kept pace with
the Fed's electronic printing press.
The second point to be understood is that bank reserves should not be counted
in the money supply (they are not available to be spent within the economy)
and are therefore not included in the TMS calculation. (Note: bank reserves
are also not included in the M1, M2, M3 and MZM calculations). An implication
is that the "bits and bytes" held in the accounts of bank customers
(which do count in the money supply) are different to the "bits and
bytes" held in reserve by the commercial banks. It's important not to
confuse the two.
The third point to be understood is that when the Fed monetises
assets under its "QE" programs it does not only add to bank
reserves; it adds to bank reserves AND deposit currency in equal dollar
amounts. For example, when a Primary Dealer sells $100M of bonds to the Fed,
the Fed adds $100M to the bank account of the Dealer and $100M to the
reserves of the Dealer's bank.
The fourth point to understand is that the inflationary effect of a dollar
added to deposit currency is the same as the inflationary effect of a dollar
added to physical currency. In fact, deposit currency is subject to
conversion on demand to physical currency. There is presently a lot more
deposit currency than physical currency in existence, but the Fed stands
ready to make up any difference between the amount of physical currency held
by the banks and the amount of physical currency demanded by bank customers.
Above are the main points we wanted to make/reiterate, but here are some
additional points to bear in mind:
a) Money Market Funds (MMFs) are not money, they are investments in
income-paying securities.
b) Time deposits are not money, they are loans to banks.
c) It could be argued that savings deposits are not money for the same reason
that time deposits are not money, but the critical difference between these
two types of deposit is that money in a savings account is available on
demand at par whereas putting money into a time deposit involves giving up
the ability to use that money for a certain period. Savings deposits are
therefore counted in the money supply.
d) The money supply cannot be affected by money going into or coming out of
the stock or bond markets, because money never goes into or comes out of any
market. For every transaction there is always a buyer and a seller, so asset
purchases involve transfers between bank accounts as opposed to money going
into or coming out of an asset. Another way to say this is that cash levels
on an economy-wide basis cannot shrink or expand as a result of rising or
falling asset prices.
e) All the cash in the economy must always be held by someone, so an increase
in the amount of cash being held does not reflect an increase in the general
desire to hold cash. It simply reflects an increase in the money supply. For
example, if 1 trillion dollars is added to the US money supply over the next
12 months, then the sum of all cash holdings in the US will have to increase
by 1 trillion dollars over the next 12 months regardless of whether the
average person/corporation wants to hold more cash. For another example, much
has been made of the increase in cash on the balance sheets of US
corporations, but the fact is that someone has to hold all the additional
cash that has been created. If not the corporations, then the general public.
f) In modern, developed economies such as that of the US, the money supply
can only increase via central bank asset monetisation
and/or the expansion of commercial bank credit, and the money supply can only
decrease via central bank asset sales and/or the contraction of commercial
bank credit and/or deposits being wiped out due to banks going bust.
g) Although uninsured deposit currency could be wiped out when a bank goes
bust, in the US this usually won't happen for two reasons. First, the sum of
the equity and debt of the large US banks is, on average, about 30% of total
assets, which means that the banks would have to suffer declines in assets of
at least 30% before uninsured depositors would start to become vulnerable.
Second, the Fed appears to be willing to underwrite everything using its
ability to create an unlimited amount of new money.
This article is excerpted from a commentary
originally posted at www.speculative-investor.com on 21st April 2013.
|
|