WILLIAM
POOLE
Brown
University
Monetary
Policy
during
the
Recession
THE DEPTH
of
the
current
recession makes
it clear ex
post that government
stabilization
policy
should have been
less
contractionary
in
1974.
In
fact,
both
monetary and fiscal
policy were extremely
contractionary not only
relative
to the needs of
a
declining economy but
also relative to policy
during the
1972-73 boom.
During 1974, the full
employment budget sur-
plus rose
sharply and
monetary expansion slowed
markedly. My task in
this report
is
to
analyze the
monetary part of recent
stabilization policies.
A
sharp
deceleration of
money growth (both
M1
and
M2) since mid-1974
is
evident in figure 1.1
Economists generally agree
that money growth in
the
second
half of
1974
was
too low, but split over the
proposition
that
money
growth
in the
first
half-6.2 percent for
M1
from December 1973
to
June
1974-was
about
right.2
As
I
argued
a
year ago,
an M1
growth target
of about
6
percent was appropriate on the
basis of
information available
in
early
and
mid-1974.3 Therefore,
I
applaud the Federal
Reserve
for
restrain-
Note:
I
want to thank
the Federal Reserve Bank
of Boston for research
support and
especially Ruth Kupfer
and Redenta de Leon of
the Bank for research assistance
and
typing.
The
views expressed,
however,
are
mine
and do not necessarily reflect those of the
Federal Reserve
Bank
of Boston.
1. I am
inclined to
believe that M2 is more closely
related to business conditions
than
M1, but have decided to
use M1 in most of my
analysis because the Federal
Reserve
tends to emphasize it
and
because the differences
in the recent behavior of the two
mea-
sures are relatively minor.
2. All
growth rates
in
this report are continuously
compounded
annual rates of
growth.
3. William
Poole,
"Reflections on U.S. Macroeconomic
Policy,"
BPEA
(1:1974),
pp.
233-46.
123
124
Brookings Papers
on
Economic
Activity,
1:1975
Figure
1.
Money
Stock
and Selected
Interest
Rates,
Monthly,
January 1971-March
1975
Billions of
dollars
(ratio scale)
Billions
of
dollars
(ratio scale)
325
300
.
600
275
-]550
275 10
MeMa,~-(right
scale) ~
250
500
225
Z
450
200
400
Percent
13
12
10
Federal
f
iunds
rate
9.
7
Yield
on Aaa
corporate
bonds
6
S
4
3
1971
1972
1973
1974
1975
Sources:
Federal
Reserve
Bulletin,
various
issues.
a. Ms
plus
time
deposits
except
large
certificates
of
deposit.
b.
Currency plus
demand
deposits
at
commercial
banks.
William
Poole 125
ing money growth
and
permitting
interest rates to
rise sharply
in the
first
half of 1974.
The next section is
devoted
to
a
defense of
the
proposition
that
a
6
per-
cent M1
growth
was
appropriate
in
1974.
There
follows
an
analysis
of
the
Federal
Reserve's explanation
of
monetary policy
in the second
half of
1974.
The
report
concludes with
a
few
general
comments on
what
the Fed-
eral Reserve should do, given
recent
experience.
Judging
the
Appropriateness of Monetary Policy
Policymaking
is
inherently
a
problem
of
decisionmaking under uncer-
tainty;
an
ex
post analysis
of
policy ought, therefore,
to be
cast
in the
same
terms. Judging
the
performance
of
policymakers requires
that
something
be
said
about
preferences
for
possible outcomes, especially
those related to
unemployment
and
inflation;
about
the
effect of events known
to
policy-
makers
when
decisions
are made on the odds of various
outcomes;
and
about the
impact on
the
odds
of
policy adjustments.
Although opinions
differ on the
relative
costs of
unemployment
and in-
flation,
I
know of no statements
by public
officials
suggesting
that the
cur-
rent
rapid
reduction in inflation has been worth the cost in terms of unem-
ployment.
On the
contrary, public officials have stressed repeatedly
that
inflation
must be reduced
slowly
in order to avoid
a
depression,
and that the
unavoidable cost
was
a
modest rise
in
unemployment.
Differences over the role of
preferences
arise
largely
because of
varying
views
of how the
world
works.
I
assign very
little
importance
to
inflation
per
se
but
feel
that
the
economic and
political dynamics
of inflation com-
bine
to
rule
out
any significant
chance of
simply stabilizing
the rate of in-
flation at 1973
or 1974 levels. Further acceleration of inflation very likely
would have generated vastly increased odds
of
deeper recession
in the
future.
I
basically agree, then, with
Milton
Friedman's argument
that the
real
choice is not between inflation and
unemployment,
but between
unemploy-
ment now
and
unemployment later.4
This is an
empirical
statement-cor-
rect or
incorrect-about the
way
the world works and not about
prefer-
ences. Nevertheless, the tradeoff between unemployment now and unem-
4.
Milton Friedman,
"The Role of
Monetary Policy," Americani Econlomic
Review,
vol.
58 (March 1968),
pp. 1-17 (especially
7-11).
126 Brookings Papers on Economic Activity, 1:1975
ployment later applies only within a certain range. After
a
point, more
unemployment now clearly brings more unemployment later, again for a
mixture of
economic and political reasons. When firms go bankrupt
and
are dismantled or fall far behind in their capital spending, an irreversible
loss occurs, and production and employment cannot be returned easily to
full employment levels. In addition, some government actions
taken in
response
to
unemployment, as
well
as
in
response
to
inflation,
reduce
eco-
nomic efficiency and interfere with the return to full employment.
I
also agree with the Modigliani-Papademos argument that the desirable
path for the economy is one involving the fastest possible recovery con-
sistent with
maintaining low odds of overshooting full employment
and
re-
accelerating inflation.5 I differ with Modigliani and Papademos not over
goals
but
over
the
odds of
achieving any given target path
for the
economy.
This
point
will be
discussed further in my comments on current policy,
but
it
is central to my analysis of Federal Reserve policy in 1974.
THE EVIDENCE OF
1974
Whatever
lessons
evolve
from the
sharp
rise
in
unemployment
in
late
1974 and
early 1975, an analysis of monetary policy
in
1974
should be
lim-
ited to the evidence available at the time the policy decisions were made.
The consensus
forecast
in
early 1974 was for
a flat
economy.
Pessimists
argued that
a
couple of quarters of declining real gross
national
product
were
likely and optimists thought a recession would be avoided. No
one
foresaw
either
a
deep recession or
a
boom.6
By June the forecasts became, if anything, a bit more optimistic. The
disruptive oil embargo had been lifted,
the
unemployment
rate
was the
same
as in
January,
and
the index of industrial
production actually
had
risen
a
bit since
April.
From
what is known
now,
the
business
cycle peak may
be
tentatively
placed at November 1973. However, after
a
significant decline
in the
early
months
of
1974,
the
economy
remained
basically
on a
plateau
until
autumn.
As
the
year continued,
the
consensus
forecast
was
revised downward,
but
5. See their paper in this issue.
6. It is fair to say that
a
year
ago many economists were concerned
over the
possi-
bility that maintaining a 6 percent
trend of money growth would
lead to a prolonged
period of economic slack, with
unemployment remaining above 6 percent,
and perhaps
rising slowly, for several years.
No one
I
know,
however, assigned any
significant
proba-
bility to
a
1975 unemployment
rate of the current magnitude.
William Poole 127
this plateau led most forecasters to believe that any further contraction
would be relatively mild. Indeed, even as late as fall, many forecasters
an-
ticipated only a moderate recession. The "Typical Quarterly Forecast for
1975"-the median quarterly forecast among twenty-three surveyed by the
Federal Reserve Bank of Richmond-had a quarterly pattern of unemploy-
ment for 1975 of
6.8, 7.0, 7.2,
and
7.3 percent, respectively.7 Since seventeen
of the twenty-three forecasts are dated December 1, 1974, or later, it is clear
that the magnitude of impending unemployment was completely unfore-
seen by the professional forecasters, or at least not foreseen with enough
confidence
to
be
included
in the
published forecasts.
Critics of
monetary policy
in
the
first half of 1974 point to three reasons
why unusually rapid money growth would
have been
appropriate: exog-
enous shocks
in the
form of the oil embargo, and short harvests in 1973;
the
sideways
movement of
the economy; and the rapid escalation of interest
rates.
Shocks.
Those who
emphasize
the
importance of the recent shocks from
farm and
fuel products have not,
as
far
as
I know, referred to similar past
episodes
in
support
of their
position favoring
a
more expansionary mone-
tary policy.
The Korean
War experience presents some intriguing parallels,
at least
on the
surface. Raw
material and
farm prices surged sharply up-
ward following
the outbreak of
the war, transferring income to foreign pro-
ducers of primary products. However, most of this income transfer
was
probably
saved
in the
short
run,
and
the
U.S.
balance on
goods
and services
(excluding military)
declined
by $4.3 billion between 1949 and 1950. Even
the
high-employment
federal
budget surplus rose until
late
in 1950.
But
the
Korean
War
shock
did not
produce
a
recession
and the
1950 acceleration
of
money growth
has
to be
viewed,
in
retrospect, with regret rather
than
pleasure.
My point
is
not to
offer the Korean
War experience
as a
counter-example,
but
rather to
argue
that the
case for a more expansionary policy in 1974
ought
to rest on more than
simply feeding
an assumed
surge
in
the
aggre-
gate price
level
into a standard
macro
model.
The
underlying micro
be-
havioral
equations
reflect the
holding of
inventories
and precautionary
balances by
individual
economic units precisely in order to cushion
the
effects of unforeseen
disturbances.
The
advocates of
a
policy response
to
the
oil
shock
ought, therefore,
to
display
some
evidence
that outsize micro
7. Federal Reserve
Bank
of Richmond, Business Forecasts, 1975 (February 1975), p.
5.
128 Brookings
Papers on Economic Activity,
1:1975
shocks in the past have
had
the macro effects postulated.
Historically,
there has been a tendency for economic policy to overreact
to external
events. Given the record, it seems best to maintain neutral
policy settings
in the absence of compelling evidence that a "clear and
present danger"
requires
a
policy response.
Sideways economy. The basically sideways movement of the
economy in
the
spring and summer of 1974 was
an
obvious danger sign.
The
pattern
was
not dissimilar, however, to that characterizing most of 1962
and
late
1966 through early 1967-periods that terminated in
renewed expansion
rather
than contraction. No solid criteria have been developed
to distinguish
the
economic "pause that refreshes" from one that forebodes
real trouble.
Interest escalation. The significance of the sharp escalation
of short-term
interest rates in early 1974 cannot be ignored even though
the
rise was
actually smaller than that in the first half of 1973. Rising
rates in 1973, as
in
most other
periods
of business
expansion,
were
appropriate
in
retrospect,
and the key question is whether sufficient information was
available to dis-
tinguish the 1974 case from the others. The economy was
obviously some-
what
weak in
1974,
but
given
the
lags
in the
effects of
a
changed
rate of
monetary growth, this weakness is relevant only insofar as it affects
the
economic
forecast for four
to
six quarters ahead. That forecast
held that
the
economy would level off,
or
fall
a
bit, and then resume its
expansion.
Interest
rates rose far more rapidly in the first half of 1974
than would
have
been
anticipated
at the
beginning of the year given
actual
money
growth.
The
proper response
of
monetary policy
to such
an
unexpected
change in interest rates depends crucially on whether the
change
is the
result
of
a
financial or
a
real disturbance. I have seen no
evidence suggesting
that it stemmed
from
a financial
disturbance.
In
U.S.
business
cycle experience, steady
or
accelerating
money growth
and rising short-term interest rates are generally associated with
expanding
economic activity; similarly, decelerating money growth
and
falling
interest
rates are standard recession
phenomena. Exceptions
are
few.
Given
these
patterns,
the
evidence
would have to be
very strong
to
prove
that the
proper
policy response
to
interest rate
pressures,
either
up
or
down,
is
a
sustained
acceleration or deceleration of
money growth.
Since
the
goal
of
monetary
policy
in
early
1974
was
to
permit
a
little slack to
develop
in the
economy
to
reduce the risk of a further acceleration in inflation,
the
Federal Reserve
appropriately
maintained
a
rate of
money growth roughly
equal
to
the
average
of the
preceding several years.
William
Poole
129
However,
monetary
policy in
the
second half
of 1974
was
ill-advised
for
exactly the
same
reasons.
Declining
money
growth in the
face of
downward
interest rate
pressures
should have
been
recognized as
characteristic of
re-
cession.
Even
ignoring
accompanying
interest
rate
movements, the
case
against
permitting
sharp
monetary
deceleration
is
overwhelming. Over
the
entire
period since
mid-1907 for
which
monthly
data
on the
money
stock
are
available, every
recession was
accompanied by
a
significant
deceleration
of
money
growth
and
every
significant
deceleration of
money
growth
was
accompanied by
recession.8
My
position
may be
summarized as
follows:
if
money
growth is
mea-
sured over
six-month
spans and
compared with
the
average rate
of
growth
over
the
preceding
two or
three
years,
most
accelerations
or
decelerations
of
2
percentage
points
or
more will
be
found to
have been
undesirable
under
almost any
preference
function.
Judgments vary
as to
the
importance
of
these
monetary
accelerations and
decelerations
in
generating
the business
cycle.
Nonetheless,
if
large
monetary
accelerations and
decelerations
are
typically
regretted
after
the fact,
then-even if
monetary
instability is
not
very
important-there is
no point
in
having
monetary
policy
push in
the
wrong
direction.
The
Federal
Reserve was
playing the
odds
correctly
in
the
first
half of
1974 by
holding
money
growth to
about 6
percent. But
by
per-
mitting the
sharp
monetary
deceleration
in the
second
half of
1974,
it was
playing
with
adverse odds
of at
least 8
to
1
calculated
from
1952-73
ex-
perience.9
The
adverse
odds were
very
much
higher if
the
extensive
prewar
recession
experience is
also
counted.
Moreover,
these
findings
were avail-
able
to the
Federal
Reserve in
the
second half
of 1974,
and
should
have
guided
policy
decisions.
There are
only
two
possible
defenses for
the
sharp
monetary deceleration
in
late
1974.
It
might be
argued,
first, that
persuasive
evidence
was
available
8.
See
William Poole,
"The
Relationship of
Monetary
Decelerations
to Business
Cycle
Peaks:
Another Look
at the
Evidence,"
Journal of
Finance, vol. 30
(forthcoming).
In
this paper the
meaning of
"significant
deceleration" is
defined
carefully.
9.
These odds
were
calculated by
measuring
money growth
over six-month
periods-
December-June,
and
June-December-and then
examining all
cases
between 1952 and
1973
in
which
money
growth was more
than 2
percentage
points below the
average for
the
preceding
three years.
The nine
periods of
sharply
decelerated money
growth were
both
halves of
1953, both
halves of
1957, the second
half of
1959, the first
half of
1960,
the
second
halves of 1966,
1969, and
1971. Given the
actual
performance of
the
economy
in
the months
following
these
decelerations, I
regard seven of
the nine
decelerations
as
clearly
inappropriate, the
1971
deceleration as
probably
inappropriate, and the
1966
deceleration as
possibly
appropriate.
130
Brookings
Papers on
Economic Activity,
1:1975
suggesting
that the
significance
of the
money
stock had
changed; or,
second,
that the Federal Reserve is
unable to control the money
stock and
therefore
is not responsible for the
sharp monetary deceleration.
These
two
argu-
ments will
now be
discussed
in turn.
Federal Reserve
Explanations
The
Federal Reserve's position
has been expressed clearly
in two recent
statements in
response to
proposed congressional legislation
concerning
the
conduct of monetary policy.10
A careful
reading
of these two statements raises questions
about
the
in-
ternal
consistency of
the Board's position. On the one
hand,
the
Board
seems to
accept
the
argument
presented in the previous section:
We appreciate the fact that
an expanding economy requires
an expanding sup-
ply of money, that any protracted
shrinkage of the money supply
may well lead
to shrinkage of economic activity,
and that attempts to encourage
growth
in
money and credit will lead to
a decline of short-term interest rates
when economic
activity
is
weak.
We
are well aware that an
expanding economy needs an expanding
supply
of
money and credit and that any
protracted shrinkage of the
money stock could
lead to or exacerbate a shrinkage
of economic activity."
On the
other hand,
the
Federal Reserve seems to argue
that lower
Ml
growth in the second half
of 1974 was acceptable because
the significance
10. In
regard to
H.R. 212,
"Statement by
Arthur F. Burns,
Chairman, Board
of Gov-
ernors
of the Federal
Reserve
System, before the
Subcommittee
on Domestic
Monetary
Policy
of the
Committee on Banking,
Currency and
Housing,
U.S. House of
Representa-
tives,
February 6,
1975," Federal
Reserve
Bulletin, vol. 61
(February 1975),
pp.
62-68
(hereafter
referred to
as
"FRB
Statement 1"); and in
regard to
Senate Concurrent Reso-
lution
18, "Statement
by Arthur F.
Burns,
Chairman, Board of
Governors of the
Federal
Reserve
System,
before the
Committee on
Banking, Housing
and Urban
Affairs, U.S.
Senate,
February 25,
1975," Federal
Reserve
Bulletin, vol. 61
(March 1975),
pp.
150-55
(hereafter
referred to
as "FRB Statement
2").
11. FRB
Statement
1, p. 63,
and
Statement 2, p. 153.
These
two statements,
however,
are less
precise than
they first
appear since
"shrinkage" of the
money stock,
interpreted
literally,
involves an
absolute decline
and
"protracted" involves a
period of
unspecified
length.
Given the
context of recent
monetary
analysis and the
frequent reference
to mone-
tary growth
rates
in
the
two
statements, it
is
reasonable to
interpret "shrinkage"
in
terms
of
a
decline
in
the
growth rate of
money. There
is no clear
interpretation of the word
"protracted," but the
Federal
Reserve has
frequently argued
that fluctuations
in
money
growth are
unimportant unless they
last for a
year or more.
William Poole
131
of
M1
has been changing. This argument, however, is
inconsistent with the
Board's
argument that
it
has attempted to achieve faster
growth in the
monetary
and credit
aggregates. Finally,
the
Federal Reserve
has argued
that it cannot control the money stock very precisely anyway.
CHANGES
IN THE
SIGNIFICANCE OF M1
The
argument that
the
significance
of M1
has changed
appears promi-
nently in both statements:
As
a rule consumers and businesses no longer hold all, or even
most, of their
spendable
funds as
currency
or
demand deposits. More and
more corporate
treasurers have learned how
to
get along with a minimum of
demand deposits; a
large part
of their transactions and
precautionary balances are
nowadays placed in
interest-bearing assets-negotiable certificates
of
deposit,
Treasury bills, com-
mercial paper, short-term municipal securities, and other forms.
Consumers, too,
have
learned
to
keep excess
funds
in
savings deposits
at
commercial
banks, shares
in
savings
and loan
associations,
certificates
of
deposit, Treasury
bills,
and
other
liquid instruments,
and
they
shift
their
liquid
resources
among
these
assets. The
result is
that no
single concept
of
money any longer
measures
adequately the
spendable
funds that are held
by
the
public.
For
example,
the
narrowly defined money
stock rose
by 41/2 per cent
during
1974.
But this
concept
of
the
money supply
has
lost
much of its
earlier
sig-
nificance.12
The
possibility
of
changing
economic
relationships
is a
constant source
of
concern
to
policymakers
and
advisers, properly
reflected
in
a
continuous
search
for evidence
of
change
and
a
willingness
to
shade
policy targets
de-
pending
on
the
strength
of the
evidence.
The
evidence
for
a
changing significance
for
M1
in 1974 is
not
strong.
Starting
with
the
April
1975 issue
of
the
Federal
Reserve
Bulletin, monthly
data
for
five
different
monetary aggregates
are
published.
The
growth
rates
for
these
aggregates
in
1973 and 1974
are
reported
in
table
1,
and
all
show
decelerations
of
roughly
similar
magnitude.
If
a
significant
shift out
of
M1
had taken
place
in
1974, divergent
movements
in
M1
and,
say, M2
would
have been
expected.
Given
the
uncertainties,
an
appropriate
policy
re-
sponse might
have
been
to aim for some deceleration
in
M1
and some ac-
celeration
in
M2,
but
in fact
the decelerations
in the
aggregates,
however
de-
fined,
were
about the
same.
12. FRB
Statement
2,
pp. 153-54.
132
Brookings Papers on Economic Activity,
1:1975
Table 1.
Growth Rates of Selected
U.S. Monetary
Aggregates, 1972-74
Annual
rate in
percent,
continuously compounded
Growth rate
Change
Monetary
December
1972-
December
1973- (percentage
aggregatea
December 1973 December 1974
points)
M1
6.0 4.6
-1.4
M2
8.5 7.0
-1.5
M3
8.5 6.5
-2.0
M4
11.0 10.1 -0.9
M5
10.1
8.6
-1.5
Source:
Calculated from dollar
figures
in
the
table, "Measures
of the
Money Stock,"
Federal Reserve
Bulletin,
vol. 61
(June 1975).
a.
The
measures
are defined as follows:
MI: Averages
of
daily figures
of
(1) demand deposits of commercial banks
other than domestic interbank
and U.S.
government deposits,
less cash items
in
process
of
collection and
Federal Reserve float; (2) foreign
demand
balances
at
Federal
Reserve banks; and (3) currency outside the
Treasury,
Federal Reserve
Banks,
and
vaults of
commercial
banks.
M2: Average
of
daily figures
for
Ml
plus savings deposits,
timne
deposits
open account,
and time certificates
other than negotiable
certificates
of
deposit of $100,000
of
large weekly
reporting banks.
M3: M2 plus deposits
at mutual
savings
banks,
savings
and
loan
shares,
and credit union shares
(nonbank
thrift).
M4: M2 plus large negotiable
certificates
of
deposit.
M5:
M3 plus large negotiable
certificates of
deposit.
The
FRB statement
above
mentions
even broader
aggregates
of
liquid
assets-including Treasury bills,
commercial
paper,
and so
forth-but
monthly
data
for
these
broader
aggregates
are
not
published. More
com-
prehensive
data are available in the
flow-of-funds
accounts,
but these are
not published
on a
timely
basis and are
of little value
in
month-by-month
policymaking.
The
rate of
turnover in demand
deposits might
also
suggest a changed
significance for M1. The
reasons
for
changes
in
turnover-the ratio of de-
mand
deposit
debits to demand
deposits-are not
well
understood; the
average
annual rate of increase of
deposit turnover between December 1964
and December
1971 was
8.8 percent while the
rates
of
increase
in
1972,
1973, and
1974
(December
to
December)
were
8.0, 19.5, and 15.0 percent,
respectively.
If these
numbers mean
anything,
the
Federal
Reserve
should
have forced
a
sharp
deceleration of M1
growth
in
1973
and a
mild accelera-
tion in 1974.
Obviously, new
theories
and new data
could overturn the empirical regu-
larity
that
sharp
decelerations of
money growth
are
associated
with
reces-
sions, but they also could confirm
the empirical regularity. In the absence
of
compelling evidence
to the
contrary,
the
safest course is to assume that
M1
has the
same
significance
as in the
past.
William Poole
133
IMPOSSIBILITY OF CLOSELY CONTROLLING
Ml
For some time the Federal Reserve
has
argued that tight control of the
monetary aggregates
is
technically impossible:
First, H.R. 212 assumes
that
the Federal Reserve
can
control
the rate of
growth
of demand deposits plus currency
in
public
circulation over
periods
as
short
as
3
months. This we are unable
to
do.
All
that we
can control over such brief periods
is the
growth
of
member
bank
reserves;
but
a
given
rate of
growth
of
reserves may
be accompanied by any of a wide range of growth
rates of the narrowly defined
money supply.'3
Clearly,
the
short-run
relationship
between
the
change in reserves and the
change
in the
money
stock has
substantial variability.
However, the Fed-
eral Reserve should be
able
to offset much of this variability. From reports
submitted
by
member
banks,
the
Federal Reserve
has a
complete
enumera-
tion
of
daily
member
bank
deposits
within eight days of
the end
of
each
statement
week, and partial
data are
available
even
sooner.
Estimates
of
nonmember
bank
data
and of
currency
in the
hands of the public
are re-
quired
to
complete the money
stock
estimates.
These
estimates
are
subject
to
errors and uncertainties,
but
the differences
between preliminary and
final estimates
are
minor
compared
to the shortfall
in
money growth
in the
second
half
of 1974.
Since
reasonably
accurate data on the
money
stock
are
available on
a
timely basis,
the
Federal
Reserve has
ample opportunity
to
react to undesired trends
in
money growth by
changing
the
growth
in
bank
reserves. Most of
the
short-run variability
in
the
ratio of money growth
to
reserve growth should
be in the
denominator
of
the
ratio-reserves-rather
than
in the
numerator-the
money
stock.
In
any case, the slowdown
in
money growth
during
1974 came
not
from a
drop
in
the
ratio of money
to
reserves
but
rather
from
a
deceleration
in re-
serve growth.
Member
bank
reserves grew
at
a
continuously compounded
rate of
11.1 percent over
the
twelve months ending
December 1973 but at
a
4.9 percent
rate
over
the twelve months ending December 1974. Moreover,
reserves grew
at
a
rate of
only 0.9 percent
in
the
six-month period ending
December
1974,
in
contrast
to an
8.8
percent
rate in
the
previous
six-month
period.
The
Federal Reserve
has
argued
that
its
attempts
to
expand
bank
re-
serves may
be ineffective
in some circumstances:
The Federal
Reserve
can
supply
the
banking system
with
reserves
through
open
market
operations
or
through
reserve
requirement
changes;
but if banks
13.
FRB
Statement
1,
p.
64.
134
Brookings
Papers
on Economic
Activity,
1:1975
choose
to
repay
debt
or rebuild their
liquidity,
these actions
will
have
little
impact
on
the
public's
money supply.'4
This position
raises questions
about two
effects of open
market operations:
the first,
on member
bank
borrowing
from the Federal
Reserve;
and the
second,
on
member bank
holdings
of excess
reserves.
Member
bank borrowing
at
the discount window
fluctuated substantially
in 1974, rising
from
a
December
1973 average
of
$1.3
billion to a
peak of
$3.4 billion
in August
1974,
and then
dropping to $0.7
billion
in
December.
The amount
of
outstanding
borrowing is,
of
course,
on the
books
of
the
Federal Reserve,
and hence
known
continuously. While
it
is true that
under
the current system
of
lagged
reserve
requirements-a
system
that fixes
in
advance
the
dollar amount
of
required
reserves
in
any particular
week-
increases
in
borrowings
cannot
be
immediately
offset
by open
market sales,
declines
in
borrowings
can nonetheless
be offset
by open
market
purchases.
The
Federal Reserve
has
argued:
"Of
late, open
market
policy
has been
reinforced
by
other
monetary
instruments.
The
discount rate was reduced
on
three occasions-in December,
January,
and
again
early this
month
[February]-from
8
per
cent to
63/4
per
cent."''5
But the
incentive for
banks
to use
the
discount
window
has declined substantially as open
market
interest rates have
fallen to
a
much
greater
extent.
The
federal funds
rate
averaged
12.01 percent
in
August
1974,
4
percentage
points above
the
dis-
count rate.
Four months
later,
in
December,
the
funds rate
averaged
8.53
percent,
or
about
3/4
of 1
percentage
point
above the discount
rate. By
March 1975
the federal
funds rate
averaged 5.54 percent,
about 1
'/4
per-
centage points
below
the
discount rate
of
63/4
percent.
The
sharp
decline
in
member
bank
borrowings
was
largely
predictable
due to this
sharp
change
in
the
spread
between
the
discount
rate and
open
market
rate.'6
For
a
given
quantity
of
total
reserves, attempts by
member banks
to
re-
build
liquidity
through
a
restructuring
of
their
assets can
contract
deposits
only
if
the banks
hold excess
reserves.
Member
bank
excess reserves
were
$262
million
in December
1973,
and
$131
million
and $339 million,
respec-
tively,
in
June
and
December
of
1974.
These small
swings
in
excess reserves
14. FRB
Statement 2, p.
152.
15. FRB
Statement 2, p.
151.
16. This
discussion
is
not
meant to
imply
that the discount
rate
should have
remained
4 percentage
points below
the
federal funds rate. The
discount
rate
is
best
held continu-
ously
above
the funds
rate;
but
having
failed
to
maintain such a
relationship
in
mid-1974,
the
Federal Reserve
should have
anticipated
the
sharp
decline in
borrowing
as
market
rates dropped
more
rapidly
than the discount rate.
William
Poole
135
are
clearly of
trivial
consequence
given
total reserves that
exceed
$35
billion.
The Federal
Reserve has also
argued
that there is a link
between
bank
loan
demand
and
money
creation:
The Federal
Reserve
can
have
a
marked influence
on
short-term
interest
rates
and
may
also have some indirect influence
on
other terms
of
credit.
But
it
cannot
force
businesses or
consumers to
borrow
from
their banks
and
thus to
expand
the
volume of bank
loans. The
Federal
Reserve
cannot force
people
to
hold
money in
the
form of demand
deposits
when
they
prefer
to
hold
their
transactions
or pre-
cautionary balances
in
income-earning
assets.'7
If
the
Federal Reserve
had maintained
a
higher
rate
of
growth of
total
member
bank
reserves,
and if member banks had
continued
their
long-
standing practice
of
holding only
minimal
amounts
of
excess
reserves
(which
earn no
interest),
then
deposit
growth
would
have
been
sustained.
Even
if
bank loan demand
had remained
sluggish,
banks would
have
pur-
chased securities
and
thus added
to
deposits
until their
liabilities
subject
to
reserve
requirements-predominantly
demand and time
deposits-had ex-
panded
sufficiently
to
absorb all of the
reserves
supplied
by
the
Federal
Reserve.
In terms of
the
composition
of
assets,
banks have
not
become
more
liquid.
Of total loans and investments for
all
commercial
banks,
loans
were
71.5
percent
at the end
of June
1974,
and
72.5 percent
at the end
of Decem-
ber.
Federal Reserve
policies
did
not
permit
banks to
increase
their
hold-
ings
of liquid
money-market
instruments.
That
banks
have
attempted, un-
successfully,
to
improve
their
liquidity
is
suggested
by
the
large
increase in
the
spread
between the
prime
rate and the commercial
paper rate.
In
June
and
July
of 1974 the
prime
rate was
only
a
little
above
the
commercial
paper
rate;
but
in
December
the
spread
was
about
1.5
percentage
points,
and it had increased somewhat further
by
March
1975.
If
the Federal
Reserve
had maintained the 1973 rate
of
growth
of bank
reserves
in
1974,
then interest rates on
open
market
securities
would
have
declined
sufficiently
to induce the nonbank
public
to
hold the
larger
quan-
tity
of demand and
time
deposits.
While
the
Federal
Reserve cannot
di-
rectly
control the
mix
of
demand
and
time
deposits,
it
certainly
can
ensure
that
sharply
reduced
rates
of
growth
do
not
occur
in
both M1
and
M2.
Finally,
the Federal Reserve
has
pointed
to
reductions in
reserve
require-
ments as
additional evidence
of
a
more
expansionary
policy:
"Reductions
17.
FRB Statement 2, p.
152.
136
Brookings Papers on Economic
Activity, 1:1975
in
member bank
reserve requirements were also
ordered-in September,
November, and January, releasing a total
of nearly $2 1/2 billion of reserves
to the banking system."18 But these
reductions
in
reserve requirements
were largely offset by open market
operations that drained a roughly
equivalent amount of reserves from
member banks. Total reserves, mea-
sured on a basis incorporating seasonal
adjustments but without adjust-
ments for
changes
in
reserve
requirements,
fell
from $36.86 billion
in
October 1974 to
$34.86 billion
in
March
1975.
Concluding Comments
The
explanation
for the
low
rate of
money growth in the second half of
1974
is,
I
feel, quite simple.
It
results
partly
from the
operating procedure
of
the Federal Reserve
that
pegs
the
federal funds rate in a
narrow band day
by day, and, given
this
procedure, partly
from
an
unwillingness
to reduce
the
peg
on
the
funds rate
rapidly enough
to
maintain
money growth.19
An enormous scale
of
open
market
operations
is
employed
to maintain
the federal funds
rate in the
narrow
band
specified by
the Federal
Open
Market
Committee.
For
example,
in December
1974,
the
net
change
in
Federal
Reserve
holdings
of
U.S. government
securities,
federal
agency
obligations,
and bankers'
acceptances was
$393 million. However, during
the month of
December, gross outright purchases
of
U.S. government
secu-
rities and federal
agency obligations
were
$1.614
billion and
gross
sales and
redemptions of these items were $432
million, for
a
net of $1.182 billion.
On
top
of these transactions were
$11.470
billion
of
gross purchases,
and
$11.895
billion
of
gross sales,
of
government
securities under
repurchase
agreements,
for
net
sales of
$425 million. Extensive
operations
in
matched
sale-purchase
transactions
("reverse repurchase
agreements")
involved
18. FRB Statement
2, p. 151.
19. The Federal
Reserve's day-by-day
operating procedure
is basically the same
as
that
used
in
the late 1960s before the adoption of targets
for
monetary aggregates.
The
continued
use of
a
procedure clearly
not well designed for the
purpose of achieving
such
targets
reflects
the
inertia
inherent
in
any organization
and
the
Federal Reserve's con-
tinuing
concern with
day-to-day
stabilization
of interest rates. This
operating procedure
has been described
in
"Numerical
Specifications
of Financial
Variables
and Their Role
in
Monetary Policy,"
Federal
Reserve Bulletin, vol. 60 (May 1974),
pp. 333-37;
and William
Poole, "The Making
of Monetary Policy:
Description and
Analysis," New
England
Economic
Review
(March/April 1975),
pp.
21-30.
William Poole
137
gross
sales
of $8.855
billion and
gross purchases of $7.962 billion, for
net
sales
of $893 million. Thus, total gross purchases for December
were
$21.046 billion, and total gross sales
and
redemptions were $21.182
billion.
Finally, transactions reported only
on
a net
basis-repurchase
agreements
of agency obligations, outright purchases
and
repurchase agreements
of
bankers' acceptances-netted $531 million.20
Open
market
activity
of this
magnitude
is
unnecessary to control member
bank
reserves. Although
Federal Reserve statements
sometimes imply that
open
market
operations
are
adjusted
so
that
the
federal
funds
rate is ex-
pected to fall in
a
certain range, Federal Reserve policy more accurately
in-
volves whatever
open
market
operations
are
required
to
peg
the
funds rate
in a
narrow day-by-day range.
The
peg is
then
adjusted occasionally
de-
pending on
the
FOMC's view
of desired
money growth and the importance
of that objective relative to the interest rate objective:
Moreover,
the condition
of
credit markets
also
weighs heavily
in
decisions
on
monetary policy. There is a school of thought that holds that the Federal
Reserve
need pay
no
attention
to interest
rates,
that the
only thing
that matters is
how this
or that monetary aggregate
is
behaving.
We
at the
Federal
Reserve
cannot
afford
the luxury
of
any such mechanical rule.21
While
policymakers
should consider
all
available
information,
the
policy
of
pegging
the federal funds
rate
day by day,
and
changing
the
peg
in re-
sponse
to
observed
money growth,
is no less
mechanical than
the
policy
of
maintaining steady money growth,
and
much
more
damaging
because it
leads to
procylical
behavior of the
money
stock.
To call
one
policy
or an-
other
"mechanical"
begs
the
issue.
Policy
must
be
based on
empirical regu-
larities,
and
whenever
relationships clearly
are
changing
it is
appropriate
to
adjust policy.
The Federal
Reserve, however,
has
offered no
evidence that
economic relationships have
been
changing
in
a
direction
that
justifies
the
policies followed
in the second
half of 1974.
These
operating procedures provide
one more reason
for
unhappiness
with
Federal
Reserve
policy
in the
first
half
of
1974-concern that it
would
20.
The December 1974
figures
discussed
in
the text
may
be found
in
Federal
Reserve
Bulletin,
vol. 61
(March
1975), p.
All.
I have not conducted
a systematic examination
of the typical gross
monthly scale of
open market operations
relative to the net
changes.
December
1974,
however,
does
not
appear to be atypical.
The
largest
net
change
for
any month
in
1974
was
August, with
a
net increase of $3.322
billion.
In
that month,
gross purchases under regular
and
reverse
repurchase agreements
alone
amounted
to $13.383
billion.
21. FRB Statement
1, p. 64.
138
Brookings Papers
on
Economic Activity,
1:1975
not permit
interest
rates to
drop rapidly
enough,
once
they
were
pushed up,
to maintain
money growth
at a
reasonable rate.
Indeed,
the Federal Re-
serve
seems
to
accept
this
argument
in the other direction as it
applies
currently:
If, for example,
we
presently
encourage
a
sharp decline
of
interest rates
on
top
of the decline
that
has
already
occurred
in
recent
months,
we would run
the
risk
of
seeing short-term
interest rates move
back up while the economy
is still receding.
There is,
moreover,
a
very
real
possibility
that,
as
a
result
of such a policy,
a
monetary base would
be
established for a new wave of inflation
in the future and
that market expectations
of
such a
development
would lead rather
promptly to a
rise
of
long-term
interest
rates.22
The fear
is
valid
under current Federal Reserve
operating
procedures.
Under
an FOMC
procedure
that
permits
only
slow and
"orderly" changes
in the
federal
funds
rate, money growth
is
likely
continually to
be
off target
as the adjustments
in the funds rate
lag
behind
market
pressures.
But
why
should
the Federal
Reserve maintain these
procedures?
No pos-
sible
gain
warrants
the costs
imposed
from
extensive
day-by-day
open
mar-
ket
operations
to limit
fluctuations
in
the federal
funds rate. Moreover, this
policy
has
obviously
failed
to stabilize
the
federal
funds rate
over spans
of
six
to twelve
months.
Excessive
money growth
over
several quarters
at
a
time-as
in 1967-68
and
1972-73-may temporarily
cushion
upward pres-
sures
on
interest
rates,
but
only
at the cost of
exacerbating inflationary
pressures
and
raising
the ultimate interest rate
peak.
Conversely,
abnor-
mally
low
money growth-as
in
1957 and late
1974-may temporarily
cushion
declines
in interest
rates,
but
only
at
the
cost
of
deepening reces-
sions.
The
Federal
Reserve
should
promptly
reform its
operating
procedures.
The FOMC should
direct
the
Open
Market
Manager to
achieve
the rate
of
growth
in nonborrowed reserves
necessary
to
attain the target
rate
of money
growth,
and
the federal
funds
rate
should
be
permitted
to
fluctuate without
limit.
The
goal
should
be
to return
promptly
to a 6
percent
money growth
path, projected
from June
1974,
in order to erase the
mistake of the second
half of 1974.
As of
April 1975,
the shortfall below
the
path
amounted to
$7.3 billion,
or
2.5 percent.
Once back on that 6
percent
path, the money
stock should
be
kept
there
until the
economy
has recovered
sufficiently
to
warrant
a
lower
trend rate of
money
growth
consistent with
long-run price
stability.
22. FRB
Statement
1, p. 64.
William Poole
139
Many economists
share the Modigliani-Papademos
position
that
a
much
more expansionary
monetary policy is
needed. I would agree
if I
were
con-
fident of model estimates
of the policies
required to engineer their
projected
employment path.
But
I
suspect that the
differences in the simulated
out-
comes under their
policy proposal
and my policy proposal
are, from past
experience, far smaller
than the model
errors themselves. The
safest course
is to
avoid major
departures from policy
settings that,
if
maintained five
years at a time, would
be expected to be
in the right range. The
recent fore-
casting fiascos
make it abundantly clear
that knowledge is
inadequate to
make
significant
departures from a trend
path of money growth
a good bet.
By trying to do
too much, policymakers
have put themselves
into a
vicious
"stop-go"
cycle
with
ever-widening
oscillations.
Each
period
of
monetary expansion
has been higher than
the previous one-considering
the
1965, 1967-68,
and
the 1972-73 expansions.
Each
of
the inflations since
1965 has been worse than the
previous
one. And
each setback
in real
ac-
tivity
since 1965 has been
deeper than
its predecessor-in
the sequence
1967, 1969-70,
1974-75. This pattern
must be broken,
and the only
method in which I have any confidence
is that of stabilizing
money growth
and
permitting the
economy
to
settle
down to a stable policy
environment.
The
discussiont
of this report is combinied with
that of the Modigliani-
Papademos report which follows.
