Are Reserve Requirements on Banks Really Needed?
William F. Hummel
In its infinite wisdom, Congress passed the Monetary Control Act of 1980
requiring the Fed's Board of Governors to impose reserve requirements of
between 8% and 14% on demand deposits in banks and other depository
institutions. In a bow to the political forces representing thousands of
small banks, the act set the required reserves on the first $25 million of
demand deposits to just 3%. In 1982, the Garn-St Germain Act went
further and exempted the first $2 million altogether. These dollar limits
were required to be raised as the total of bank deposits grew. As of 1995
the fully exempt part had been raised to $4.3 million and the 3% reserve
part had been raised to $52 million. Above $52 million, the Fed imposes
a reserve requirement of 10%. Savings accounts, which are the largest
component of bank deposits, are fully exempt from reserves.
I present this information by way of background to ask the question
whether reserve requirements are needed at all. So far as I can judge,
the only role that the requirements now play is to give the Fed a lever
for controlling the interest rate on overnight lending between banks, i.e.
the Fed funds rate. By forcing banks to comply on a bi-weekly basis with
the reserve requirements, the Fed creates an active money market among
banks whose price it can control through its open market operations. The
Fed discourages a bank borrowing at the Discount Window until it has
exhausted other options. However it does stand ready to lend if a bank
finds it necessary in order to meet its reserve requirements. But is this
complex system really necessary?
The system works well enough so there is no obvious need to change it.
However it does require a great deal of information and skill on the part
of the Fed as well as a lot of work by the banks in anticipating their
reserve positions. Banks don't like to hold reserves in excess of what is
needed to safely cover their daily balances. Reserves earn no interest.
By the Fed's own admission, reserves now play a relatively limited role in
controlling the money supply. Reservable deposits represent a rather
small fraction of the effective money supply. Furthermore, banks have
developed ways of reducing the apparent amount of such deposits, as for
example the use of overnight sweep accounts that convert demand deposits
into MMDAs, not subject to reserves. What limits banks from excessive
lending is not the reserve requirement, rather it is the equity/asset
ratio requirement. An insolvent bank will lose its charter. One whose
equity/asset ratio is too low will be subject to supervision of its
lending by its regulating agency. That indignity happened to Citibank
just a few years ago.
So the question is -- can the Fed control short term interest rates
effectively in the absence of a reserve requirement on demand deposits?
It is my understanding that the Bank of England and the Bundesbank do not
impose reserve requirements, and they seem to have interests rates and
bank lending under control. What if the Fed provided loans on demand to
any qualified borrower at its target rate, while still requiring adequate
collateral? This would clearly set an upper limit on the price of money,
namely the Fed lending rate. The lower limit would appear to be a pure
market rate. Would banks willingly lend their surplus funds (in excess of
equity requirements) at a rate below the Fed rate? I think not, so long
as demand for commercial and consumer loans was adequate. Of course,
one thing the Fed would have to forbid, as it does now, is an arbitrage
game -- bank borrowing low from the Fed to loan at a higher rate to its
customers.
No doubt there will be some who argue that reserve requirements are
necessary in order to prevent a runaway condition of bank-created money
leading to a general price inflation. Certainly if the cost of borrowing
is too low, this could happen. But that would occur only if the demand
were too low or the supply of available funds were too high. If demand
were too low, the problem would be self-healing. If funds were too
plentiful, the Fed could always soak them up with open market sales of
Treasury securities, even if it meant accepting a below market price to
close the deal.
From this brief analysis, I lean towards the opinion that the present U.S.
system of monetary control is unnecessarily complex. I would welcome the
views of others.
William F. Hummel
Steven Hales
In article <33714fe2...@news.pacbell.net>, wfhu...@pacbell.net
says...
> Subject: Are Reserve Requirements on Banks Really Needed?
> From: wfhu...@pacbell.net (William F. Hummel)
> Newsgroups: sci.econ
In October 1979, it (The Fed) began targeting the quantity of reserves-
specifically, nonborrowed reserves. A predetermined target path for
nonborrowed reserves was based on the FOMC's objectives for the M1 money
stock. If M1 grew faster than the FOMC prescribed, actual required
reserves would grow faster than nonborrowed reserves; the faster growth
of required reserves, in turn, would produce upward pressure on the
federal funds rate and other short-term interest rates. The rise in
interest rates would then reduce the amount of M1 deposits demanded
by the public, and M1 would be brought back toward its targeted path.
Later, however, the combination of interest rate deregulation and
financial innovation disrupted the historical relationships between
M1 and the objectives of monetary policy. In response, the Federal
Reserve in late 1982 shifted from controlling M1 through a reserves
oriented approach and returned to accommodating short-run fluctuations in
reserves demand and preventing these fluctuations from affecting the
federal funds rate. At the same time, Federal Reserve policy decisions
became conditioned on a much wider range of economic and financial
variables, including M2 and other broad monetary and credit aggregates,
that seemed more closely linked than M1 to the long-term goals of
monetary policy. Since 1982, daily open market operations have been keyed
once again to achieving a particular degree of tightness or ease in
reserve market conditions rather than to the quantity of reserves
outstanding.
(source Federal Reserve) http://www.bog.frb.fed.us/pf/pdf/frspurp.pdf
I think that your objection to complexity stems from the 1979-1982 period
when the Fed was targeting quantity of reserves and not being
accomodative.
Since deregulation specifically the DIDMCA of 1980 the Fed expanded the
total number of banks, depository institutions, under its aegis. And
therefore its control over the M1 aggregate. But deregulation also
weakened Fed power and its focus shifted directly to the Fed Funds Rate
and the Discount Rate. Since M2 and M3 dwarfed M1 the money supply grew
almost independent of Fed action on reserves. However, I am reluctant to
allow the Fed to give up any of its tools fractional reserves, bank
balance sheet oversight to control the money supply either directly or
indirectly. Bankers can be awfully dumb remember the S&L crisis and the
rash of bank failures.
Actually I favor reserve requirements on time deposits as well and Fed
control over money market accounts at investment firms.
Regards,
Steven Hales
MarketPath
"May your MC always equal your MR. But we're never that bright or we
wouldn't need designated drivers."SH