Full text of Review (Federal Reserve Bank of Dallas) : December 1976

View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

Rlderal Reserve Bank of Dallas

Business Review

.r- .
I •

•, '
"II

i\,

'1"\ I

·1"

....

""
I

'.1' -":~~,,II\ " '"
,'''. \' "tit
. '" • ~\;.'
\~I'II"
4'-I '.' 'l;
,-,
....

December 1976
Real Estate FinanceAdvantages of Innovations
In Variable-Rate Mortgages

,
,111\.'1 . ~.'~ ~ I' •.

Reserve RequirementsStructure an Impediment
To Monetary Control?
This publication was digitized and made available by the Federal Reserve Bank of Dallas' Historical Library (FedHistory@dal.frb.org)

Real Estate Finance-

Advantages of Innovations
In Variable-Rate Mortgages
Rising inflation in the United
States, because of its effect on
mortgage markets, has tended to
obstruct the allocation of resources
to residential housing, One reason
is the impact of expected inflation
on the cost to borrowers of new
mortgages. When inflation is
widely anticipated, interest rates
include a premium to compensate
lenders for an expected reduction
in the purchasing power of their
investment.
The resulting increase in mort;.
gage payments creates a financing
gap to the extent that family
income does not keep pace with
the higher interest payments in
the early years of the mortgage.
Even though inflation increases
family income relative to payments in the later years of the
mortgage, the financing gap may
force many households to downgrade their housing purchases or
even to forego homeownership.
Another reason why inflation
may interfere with the allocation
of resources to housing is the
impact of realized but unanticipated inflation on the supply of
mortgage credit. Most mortgage
funds are provided by specialized
institutions that obtain their
loanable funds from short-term
deposits. By contrast, a significant
portion of their portfolios is comprised of long-term loans. In a
period of unanticipated inflation,
the yields on these loans tend to
fall below current short-term interest rates.
Consequently, the return on the
portfolios of these mortgage spe-

cialists becomes too small to allow
them to compete for short-term
deposits with other institutions
making shorter-term investments.
Mortgage specialists have remained
solvent during rising indation in
the United States largely because
interest ceiling regulations prevented them from competing for
deposits on the basis of yields. But
as a result of these regulations, the
supply of mortgage credit has been
restricted, especially in periods
when short-term rates were at
cyclical highs.
Adoption of the variable-rate
mortgage (VRM) is one of the
major proposals of recent years to
improve the functioning of mortgage markets in an inflationary
environment. In 1975, the Federal
Home Loan Bank Board proposed
a change in its regulations to permit federally chartered savings
and loan associations to offer
VRM's on a broad scale.' However,
because of objections by consumer
and labor groups and substantial
congressional opposition, the board
withdrew its proposal on singlefamily structures but allows
VRM's on multifamily structures
and commercial real estate. Nevertheless, alternative kinds of more
flexible mortgages for homeowners
are under study by Congress.
The benefits of VRM's could be
enhanced if innovations are introduced into the loan design to
reduce the financing gap. The
resulting loan plan is called the
constant-payment-factor variablerate mortgage (CPFVRM). While
the CPFVRM has certain draw-

backs, it offers a combination of
desirable qualities not available in
other mortgage designs, which
facilitate financing of family housing in an inflationary environment.
Standard variable-rate mortgage
The VRM allows the mortgage rate
to change over the term of the loan
by indexing it to an appropriate
reference rate. Changes in the
mortgage rate can be implemented
through two methods. First, the
monthly payment may be allowed
to vary, holding the term of the
loan fixed. Alternatively, the
monthly payment may be fixed,
and the teno allowed to vary. But
the second method is not very
practical since increases in interest
rates can lead to extremely long
maturities. Consequently, recent
proposals for VRM's in the United
States center on plans with fixed
maturities.

The principal advantage of
the variable-rate mortgage
over the fixed-payment mortgage is that interest revenue
on all outstanding loans can
roughly vary with the current
cost of funds.
The principal advantage of the
VRM over the fixed-payment
mortgage (FPM) is that interest
revenue on all outstanding loans
can roughly vary with the current
cost of funds. Thus, mortgage specialists are in a better position to
maintain adequate profit margins

1. The Federal Home Loan Bank Board had imposed regulations in 1972 to prohibit federally chartered associations from
issuing mortgages on which interest is increased by adjusting payments above their initial level. Increases in interest
were allowed by extending the loan maturity-but only up to 30 years. Most state-chartered financial institutions found it
impractical to issue variable-rate mortgages became of state usury laws.
Business Review I December 1976

and can compete more readily for
deposits with diversified lenders
even though realized inflation may
not have been anticipated. Since
this provides a more stable supply
of funds to finance housing, both
lenders and borrowers can benefit.
And those that borrow when interest rates are cyclically high avoid
becoming locked in to high mortgage costs over the entire term of
a loan.
Nevertheless, many consumer
and labor groups in the United
States remain opposed to the
adoption of VRM's. Some fear the
adoption of VRM's will actually
restrict, rather than increase, the
availability of mortgage credit.
They contend that as interest
rates and payments rise on VRM's,
so will the proportion of defaulted
loans. Mortgage lenders would
then adopt more restrictive lending practices-particularly for lower
and middle-income groups.
Others oppose VRM's on the
ground that the most important
problem in housing markets is high
mortgage payments, caused largely
by the financing gap, which cause
many families to forgo homeownership. They point out that VRM's
provide no solution to this problem and suggest efforts should be
devoted to other reforms.
Experiences with VRM's
The :first major market test for
VRM's in the United States was
in California in 1975. Variable-rate
mortgages had been made by some
small savings and loan associations in California ever since 1961.

But it was not until the first half
of 1975 that six large state-chartered associations, holding over 30
percent of the industry assets in
California, began offering VRM's
on a broad scale. 2
To date, a substantial volume
of such loans has been made. For
example, in the last nine months
of 1975, two-thirds of all new
mortgages issued by the six associations were variable-rate mortgages. Since then, additional statechartered associations and two
large commercial banks have also
begun making VRM's. However,
it is too early to judge the impact
of variable-rate financing on the
availability of mortgage credit in
California or on the demand for
family-owned housing.
On the other hand, variable-rate
mortgages have been the major
instrument for residential real
estate finance in the United
Kingdom and Canada for several
years.' Their experiences can help
in assessing the advantages of
adopting variable-rate financing
in the United States since mortgage markets and institutions in
the three countries are somewhat
similar and inflation has become a
growing problem for all three.
In the United Kingdom, 90 percent or more of new mortgage
credit for owner-occupied dwellings is normally provided by building societies. These institutions
have asset-liability structures comparable to those of savings and
loan associations in the United
States. For example, over 80 percent of their assets are held in

long-term mortgages, and close
to 90 percent of their funds are
raised through savings shares on
which withdrawals can be made
on short notice.
Building societies began offering
variable-rate mortgages in the
1930's, By 1967, over 80 percent
of their loans were in this form,
And today, almost all new housing
loans in the United Kingdom have
variable rates.
The typical variable-rate residential mortgage in the United
Kingdom is issued for a term
between 20 and 25 years on a
fully amortized basis, Changes in
the mortgage rate can be accomplished by adjusting either the
payment level or the loan maturity. Before 1969, they were
achieved ahnost exclusively
through variations in maturities, but rising interest rates
created situations in which the
original payment level on many
older loans was not sufficient to
cover interest costs. So today,
rate changes are typically made
through adjustments in the mortgage payment.
The interest on variable-rate
mortgages in the United Kingdom
is not linked to a market-determined reference rate; instead, rate
changes are left to the discretion of
the lender. A practice has evolved
since 1940 whereby the mortgage
rate, as well as the rate paid on
savings deposits, is determined by
the recommendation of the Council of the Building Societies Association-the major trade association
for these mortgage lenders. There

2. The recent California experiment with variable·rate mortgages is described in Mark J. Riedy, "VRM's in California: The
Early E:zperience," Federal Home Loan Bank Board Journal 9, no. 3 (March 1976). Savings and loan aS9OCiations in
California, as well as in Wisconsin, issued variable·rate mortgages on a small scale for many years before 1975. And the
Fann Credit System has experimented with them nationwide since 1970. These experiences are discussed in Eleanor
Erdevig, "Is There a Future for Variable Rate Mortgages?," Business Conditio1l8, Federal Reserve Bank of Chicago,
November 1975; and James A. Millar and Stanley R. Stansell, "Variable Rate Mortgage Experience of the Farm Credit
System," Financial Management 4, no. 4 (Winter 1975).
3. The use of variable rate mortgages in the United Kingdom and Canada is described more fully in David L. Cohen and
Donald R. Lessard, "Experience with Variable·Rate Mortgages: The Case of the United Kingdom," and Dona1d R.
Lessard, "Roll·Over Mortgages in Canada," both in New Mortgage Designs for Stable Housing in an Inflationary
Environment, Federal Reserve Bank of Boston Conference Series, no. 14, January 1975. See also Jack Revell, Flexibility
in Housing Finance (Paris, Organization for Economic Cooperation and Development, 1975).

are no legal ceilings on mortgage
rates or savings deposit rates. So,
the arrangement could be capable
of allowing flexible adjustments in
mortgage rates that give building
societies adequate profit margins,
allowing competitive yields on savings deposits, and providing a more
stable supply of mortgage credit.
But partly because of political
pressures, the Building Societies
Association has been reluctant to
raise mortgage and deposit rates
until deposit outflows were experienced by lending institutions. A
mechanism to reduce the burden
of the financing gap--and, hence,
reduce the political pressure to
keep mortgage rates from risingwould have helped the use of
variable-rate mortgages to be
more effective in stabilizing the
supply of mortgage credit in the
United Kingdom.
Most mortgage credit in Canada
is provided by four types of financial institutions. Two of these,
mortgage loan companies and trust
companies, specialize in housing
finance. Trust companies fund
their investments through issuance
of time deposits maturing within
five years and with in1lows to
demand deposit accounts. The liability structure of mortgage loan
companies is similar, but it includes
time deposits and debentures with
maturities that exceed five years.
The other institutions, chartered
banks and life insurance companies, are diversified lenders; and
together, they hold approximately
44 percent of total mortgage debt.
A rollover variety of the variable-rate mortgage bas been offered
in Canada on conventional loansthose that carry no government
guarantee-since 1931. But until
1969, only standard fixed-payment
mortgages were permitted on loans
guaranteed by the federal government. Then, the law was changed
to allow five-year rollover contracts on these loans as well, provided their original maturity
Busineu Review I December 1976

Mortgage Terms in Period of Less Than Fully Anticipated Inflation
- - STANDARD FIXED·PAYMENT MORTGAGE (FPM)
- - - . VARIABLE·RATE MORTGAGE (VRM)
.. ......... CONSTANT.PAYMENT-FACTOR VARIABLE·RATE MORTGAGE (CPFVRM)
1,200 DOLLARS - - - - - - - - - - - - - - - - - - - -

900

.;,;

600

,'.

;;,..- ..." l....,"'~'" ..............................
......' .'. '.
_ './ " .;'"
.... .... .... -. ... .... .

...,..

-- _

300

INTEREST CHARGE

'.... :::- ...........
---"~'"

O~---,-----,-----,r----.r---~r2.000 DOLLARS

- - - - - - - - - - - - - - - - --,=-

1,700

NOMINAL PAYMENTS

1,400
/

1100

W!:::.::.~v·:(·:'

.... -

~/.

;.~

•••.••••••

" .... "....-- ~-~... ..." ....

-.;

•••••••••••••••••... --•••

-------

" ...... ..

600

-.-------r---,.------,----r---..---

1,600 DOLLARS - - - - - - - - - - - - - - - - - - - -

'REAL PAYMENTS
( 1967 DOLLARS)

600
400

I' .... ~.-::-: •.,... _ _ _ _ ""' :;.':::::::.~ ............." •••••••••

..."...... iI

1.200

-- , -....

-r------,r----.---~---_r---_r-

32,000 DOLLARS - - - - - - - - - - - - - - - - - - - -

24,000 -

.... "" ........

16,000 UNPAID PRINCIPAL

--<~:.=~:.=.::::.~....

8,000 -

.... ,".'.

.•......

0 ~-----.-----.----_,----_,----_1--1950

1955

1960

1965

1970

1975

NOTE: Terma e re eeleulated on e .. mlannual beaia lor a $30.000 loan with a 25'1ear te,m.
Initial paymenta 10' the FPM and tha VRM are caleulated at a cont,act ,ate 01 4.6
percent. The- Initial payment lor the- CPFVRM II c:atcutated at a ,ete of 4 perc:ent.
The- debiting fector for both v.rlable.rete mortgege designs ia computed b1 linking
the initllli eontract ,ete 01 4.6 perc:ent to the yield on one.yea, U.s. Gove,nment
.ecu,III... Pe1mentt 10' the- CPFVRM e .. liMed to eheng .. In the eonaume, p,lc:e
Inde.. The eonaLlme, prlee index il . Iao uaecl to d.lI.te nominal poeymentl for .11
th..e mortg.ge dealgnl to obl,ln r.. l veluel.

exceeds 24 years. By 1973, practically all new mortgages on singlefamily dwellings were of the rollover type.
Rollover mortgages are issued
with a fued interest rate and a
fu.ed schedule of payments for
each five-year term. The payment
is calculated to fully amortize the
loan over its remaining term to
maturity, which originally ranges
from 20 to 30 years for conventional mortgages and 25 to 40
years for government-guaranteed
mortgages.
At the end of each five-year term,
the outstanding balance comes
due, but the holder of either a
conventional or government-guaranteed mortgage has the option of
refinancing it at the current market rate, usually without additional
closing costs. A borrower with a
government-guaranteed loan has
the additional option of increasing
the maturity of his mortgage up
to 40 years to prevent payments
from increasing when interest
rates rise.
Variable-rate financing has
helped stabilize the supply of
mortgage credit in Canada. Specialized mortgage lenders began
increasing the length of their liabilities in the 1960's to match
more closely the five-year maturity
of rollover mortgages, and by 1969,
all statutory ceilings on mortgages
and deposits were abolished. Mortgage lenders have, consequently,
been able to maintain satisfactory
profit margins in recent years. As
a result, the severe disintermediation that occurred in mortgage

ing can be enhanced by a mortmarkets in the United States in
1974 appears to have been avoided gage design that provides not
only a competitive return but
in Canada, even though market
also a time profile of scheduled
pressures pushed interest rates
payments more similar to that
up sharply.
of money incomes-and, hence,
Still, housing starts in Canada
diminishes the financing gap. A
experienced a sharp decline in
1974. Observers suggest that high new mortgage design with these
features is called the constantpayments on new mortgages
payment-factor variable-rate mortreduced the amount of singlegage (CPFVRM).'
family housing demanded. In the
The CPFVRM has two char1963-73 period, for example, payacteristics that distinguish it from
ments on a typical mortgage for
other fixed-term mortgages. o First,
a 25-year tenn increased 50 perpayments can be designed to start
cent faster than the overall consumer price index-mainly because low and increase over time with
prices and family incomes. Second,
of higher mortgage rates. Then,
in 1974, as mortgage rates jumped separate interest rates are used to
compute payment levels and inter250 basis points above their 1973
est charges.
average, monthly payments on a
The interest rate used to comnew mortgage rose twice as fast
as the consumer price index. High pute payment levels is called the
constant-payment factor. The
mortgage costs undoubtedly
initial payment is determined by
caused some households to withamortizing the loan's principal
draw from housing markets.
Observers of the Canadian econ- over its full term at this constantpayment factor. Ideally, the payomy also suggest that high mortgage costs reduced activity in mar- ment factor can be chosen to
approximate the "real"-or inflakets for multifamily housing.
Apartments are generally financed tion-free.-mortgage rate so that the
with standard fued-payment mort- first payment is at a level that
eliminates the financing gap.
gages in Canada. Many investors
Interest changes are determined
avoided new apartment projects in
in the same manner as with the
1974 because of uncertainties
about whether future rents would standard VRM, by applying a
compensate for high fued interest current interest rate, or debiting
costs that reflected, in part, expec- factor, to the unpaid principal.
Periodic adjustments in the debittations of continued inflation.
ing rate are made by indexing it
Constant-payment-factor
to an appropriate reference rate
variable-rate mortgages
that reflects the cost of funds to
The experiences in Canada and the mortgage lenders.
Monthly payments are recomUnited Kingdom suggest that
benefits from variable-rate financ- puted at the same time as adjust-

4. The COJ18tant-payment·factor variable-rate mortgage and other types of loans with low·start payments are discussed in
Richard A. Cohn and Stanley Fischer, "Alternative Mortgage Designs," New Mortgage Designs lor Stable Housing in
an Inflationary Environment. Federal Reserve Bank of Boston Conference Series, no. 14, January 1975. One of these
designs, the graduated-payment mortgage (GPM). is geared for payments to increase on a prescribed schedule determined
at the time the loan is made. It is currently bein g studied by Congress and will be insured under an experimental program
by the Department of Housing and Urban Development in late 1976.
Since payment changes are scheduled in advance, the GPM exposes both borrower and lender to risks {rom unanticipated inflation. Higher than anticipated inflation will reduce the real return to the lender, and lower than expected
inflation will increase the real costs to the borrower. The CPFVRM. on the other hand, includes provisions designed to
eliminate both of these risks.
5. The formulas used to calculate payment level a nd unpaid principal for the FPM. VRM. and CPFVRM are discussed in
the accompanying technical appendix .

•

ments are made in the debiting
rate. This is done by amortizing
the principal outstanding at the
end of the last adjustment period
over the remaining term to matu·
rity at an interest rate equal to the
constant·payment factor. The
resulting payment level is then
increased in proportion to the infla·
tion that has occurred since the
previous adjustment to arrive at
the actual monthly payment.
Benefits from variable-rate
financing can be enhanced by
a mortgage design that provides not only a competitive
return but also a time profile
of scheduled payments more
similar to that of money
incomes-and, hence, diminishes the financing gap.

The paths for payments and
the outstanding principal are
determined by the actual course
for infiation and the difference
between the debiting rate and
the payment factor. Suppose, for
example, that a CPFVRM is issued
when no inflation is anticipated
and that its debiting rate and pay·
ment factor are identical. Further·
more, suppose no inflation develops
so that the two rates stay appron·
mately equal. f
In this case, the CPFVRM will
behave much like an FPM. The
nominal value of its payments will
remain at about the initial level
and, since prices are stable, so
will the real value, or purchasing
power. In addition, the amount of
its outstanding principal will
decline throughout the loan's term
at approximately the same pace as
with a comparable FPM.
On the other hand, suppose an
inflationary environment has

pushed the debiting rate above the
payment factor. After recomputa·
tion, the first payment will be the
same as before, but a higher por·
tion of it is required to cover inter·
est charges than would be the case
with an FPM at an interest rate
equal to the payment factor. Con·
sequently, in subsequent periods,
amortization of the outstanding
principal over the remaining term
at an interest rate equal to the
payment factor will tend to yield
higher levels of payments.
So long as the difference between
the two interest rates approxi·
mately averages out over time to
the inflation rate, nominal payments will increase at about the
same rate as prices. And the real
values for payments and the outstanding principal will be about
the same throughout the term of
the loan as those for a compa·
rable FPM in a noninflationary
environment.
Pattern of payments of CPFVRM
To illustrate the advanages of
the CPFVRM in an inflationary
environment, the path for its interest charges, payments, and unpaid
principal is compared with the
paths of an FPM and a VRM over
the postwar period. Three loans
for $30,000 with a 25-year matu·
rity are assumed to have been
issued in 1951.
The first is an FPM made at
the contract rate of 4.6 percent
then prevailing on conventional
mortgages. The second, a VRM,
is assumed to have been issued at
the same initial contract rate but
with subsequent payments recomputed every six months by indexing interest charges to movements
in the yield on one-year U.S. Gov.
emment securities. This same vari·
able rate is used as the debiting
factor in calculating interest

charges on the CPFVRM. The
constant-payment factor for the
CPFVRM, chosen to approximate
the real rate of interest, is fixed at
4 percent, and the consumer price
index is used to link payment levels
to changes in prices.
Interest rates rose over the
25·year period, increasing the debiting factor on the VRM and
CPFVRM from 4.6 percent to
11.B percent by 1974. Consequently, total interest earned on
the FPM is 30 percent less than
with the VRM and 37 percent less
than with the CPFVRM.
For the lender, the advantage
in interest earnings from the
CPFVRM is offset, to some degree,
by the fact that payments are
lower than for the other mort;..
gages in the first few years. For an
organization building a new port·
folio in CPFVRM's, this presents
a cash flow problem that ini·
tially reduces the funds available
for reinvestment. Nevertheless,
a portfolio comprised of either
VRM's or CPFVRM's, rather
than FPM's, would have offered
mortgage lenders a clear advan·
tage in earning power.
The initial payment level on the
CPFVRM would have been 2.5
percent lower than on the other
mortgages. The advantage to
the borrower is relatively small
because, in 1951, inflation had not
yet created a large financing gap.
However, if the experiment begins
with 1970, a contract rate of B.3
percent is substituted in the computations for the FPM and VRM.
and initial payments would have
been 31 percent lower with the
CPFVRM.
The advantage of a CPFVRM
to the borrower in reducing the
financing gap is offset, to some
extent, by uncertainties surround·
ing the future course of payments.

6. It is assumed here that movements in the debiting rate result primarily from changes in in1l.ationary ell:pectations. This
is a reasonable assumption !lince economic !ltudies show that the real rate of interest hu remained relatively unchanged
in the United States.

Buineu Review I December 1976

Mortgage Terms in Period of More Fully Anticipated Inflation
- - STANDARD FIXED-PAYMENT MORTGAGE (FPM )
- - _ . VARIABLE-RATE MORTGAGE (VRM)
.......•••. CONSTANT-PAYMENT-FACTOR VARIABLE -RATE MORTGAGE (CPFVRM)
1.5 THOUSAND DOLLARS - - -

1.7 THOUSAND DOLLARS - - -

experiment beginning with 1951,
the largest one-period increase in
real payments is only 1 percent,
compared with 15 percent for
the VRM.
Summary and conclusions

The standard fixed-payment
mortgage has been the dominant
f.:
j ' 0:\
I '
1.3
1.1 instrument
used in real estate
\
I
\
....
.
f,
I
\ \ ....
finance
in
the
United States since
.........
\'
I
\
......
f ..
I
....
.9 1.1
..
'
the
late
1930's.
Over much of
......... ..... .. '
-''''':..:::/
the period, it has hmctioned
.9
admirably to provide economical
.7 INTEREST CHARGE
NOMINAL PAYMENTS
financing for housing through the
.7 - , - . - - - . , . - . , . - - , - - , . - , .5 -,-,r-r-r-...-,.--.existing system of mortgage specialists. However, since the mid1.4 THOUSAND DOLLARS - - 31 THOUSAND DOLLARS - - sixties, inflation has interfered
with the ability of the fixed30 - ..............
............... .
1.2
payment mortgage to serve the
.... "
..........
needs of lenders and borrowers.
1.0
29 "
When inflation is unanticipated,
,
......
it
reduces the profitability of
....
...................
.828 ' ................
portfolios in fixed-payment mort·
'" .....
UNPAID
"gages, and mortgage specialists
REAL PAYMENTS
.6
27 PRINCIPAL
' ....
(1967 DOLLARS)
cannot effectively compete for
short-term funds. Moreover,
.4 -'-'---TI-.-rl-,.--,26 -'I-'---TI-'-'-"'-'Iexpectations of continued infla'69
"71
' 73
"75
'69
"71
"73
'75
tion increase interest rates on
NOTE: Initial paymenls for Ihe FPM and Ihe VRM are calculaled al a conlracl .ale of 8 .3
new mortgages. The resulting
pe .ce nl. Olhe f d ela ;ls 01 Ih e calcula tio ns a, e Ihe same a s Ihose fOf lhe 1951.15
increase
in mortgage payments
pefiod.
creates a financing gap, which
causes some households to forgo
homeownership.
The variable-rate mortgage has
been
proposed as a means of
If, by chance. a family's income
smaller payment increases are
does not keep pace with inflation, required later to amortize the out- improving the operation of mortpressure would develop on the
standing balance. Of course, more gage markets in an inflationary
household budget, perhaps causof the financing gap remains when environment. Since the standard
ing delays in payments or even
a higher payment factor is selected. variable-rate mortgage can increase
default. Also, short-run changes
But in most cases, a payment fac- the ability of mortgage lenders to
compete for funds, it can help stain payments caused by fluctuations tor can be chosen that eliminates
in market rates of interest might
at least part of the gap and, at the bilize the supply of mortgage
credit. However, the benefits to a
cause problems. But these dissame time, is consistent with a
borrower can be enhanced by also
advantages may by no means be
degree of risk acceptable to borincluding a provision in the mort·
overwhelming in many cases.
rower and lender.
gage that reduces the financing
The risks of default with the
Moreover, short-run changes in
gap. The resulting loan plan,
CPFVRM because a family 's
payments are not likely to be as
which has yet to be tested in the
income may not keep pace with
large with a CPFVRM as with a
United States to any great extent,
inflation can be reduced by select- VRM. The path for payments
is a constant-payment-factor variing a higher payment factor. In
is smoothed in the case of a
this fashion, initial payments are
CPFVRM since the same payment able-rate mortgage.
made higher, and the mortgage
The constant-payment· factor
factor is employed to amortize
principal is repaid faster in the
variable-rate mortgage allows payprincipal throughout the term of
early tenn of the loan. Therefore,
ments to start at a relatively low
the loan. For example, in the
1.3

F
~

~ \~

\ \

1.5 -

,,-

'-'

,
'

level and increase over the term
of the loan at approximately the
same rate as prices. Therefore, the
purchasing power, or real value,
of the payments remains fixedjust as for a standard mortgage in
the absence of inflation.
There is uncertainty, which does
not WElt with the standard fixed-

payment mortgage, about whether
a family's income will keep pace
with the rising nominal value of its
mortgage payments. Therefore,
some families and lenders may well
prefer fixed-payment financing.
But as a means of both reducing
the financing gap and providing a
satisfactory profit margin, the

constant-payment-factor variablerate mortgage would be advantageous to many borrowers and
lenders.

-William R. McDonough

Technical appendix
The standard fixed-payment mortgage is designed so that a constant-payment level
will provide the agreed-on nominal rate of
return and fully amortize the original principal over the term of the mortgage. The unpaid principal, or par value, of the mortgage
can thus be expressed as:
(1) M ,_1 =

•I P(1 + R)'--', t = 1, ... , T,

where M is the unpaid principal, t is a subscript for time, T is the loan's term, P is the
mortgage payment, and R is the contract
rate of interest.
The standard annuity formula is applied
to (1) to obtain a solution for the unpaid
principal:
(2)
= P (1- (1 R)~T-'J R-'.
Equation (2) can then be solved to determine the payment level in terms of M o, the
value of the principal at the time the mortgage is issued:
(3)
P = R M, [1- (1
R) -TJ -'.
An alternative fonn for expressing the
value of the unpaid principal-M rthat is
useful in comparing the features of the standard med-payment mortgage with those
of other mortgage forms can be solved from
(1) through (3), as:

M,_.

(4) M,

M , (1

+ R)' - P

,-.
,

i= O

+ R)',

t = 1, . .. , T.
The first term on the right-hand side of (4)
represents the rate at which interest is
being debited, or charged, to the loan. The

Bwmeu Review I December 1976

second term represents the rate at which
interest and principal are being credited.
The standard variable-rate mortgage is
designed to fully amortize its principal over
the term of the loan. However, it differs
from the standard fixed-payment mortgage
in that its contract rate, RVRM, is allowed
to vary with changes in market conditions:
(5)
RVRM, = RVRM 1 RFI-RF ,_h
t = 2•.. . , T,
where RF is an appropriate reference reflecting the cost of funds to mortgage lenders. If the maturity of the loan is fixed,
payments will vary with changes in RVRM.
The current payment level. PVRM" is
determined by using the standard annuity
formula to fully amortize the unpaid principal, MVRM ,_1 , over its remaining term:
(6) PVRM, = RVRM, X MVRM,_.
X [1- (1
RVRMt),-:r-l]-I,
t=l, ... ,T- l
PVRM, = MVRMH (1 + RVRM,) ,
t = T.
The value of the unpaid principal depends
on the pattern of previous interest rates
and payments:

(7) MRVM,

= MVRM, n
;= 1

+ RVRM,)

+ ;TIPVRMt
(1 + RVRM.)J},

-{PVRMr

X[ :.

1< _ 1+ 1

t = 1, . . .• T.
The constant-payment-factor variablerate mortgage differs in that separate inter--

est rates are used in computing payments
and interest charges. Ideally, the interest
rate used to compute payments-the constant-payment factor. or CPF-is chosen to
approximate the real. or inflation-free, rate
of interest 80 that the financing gap is eliminated. The current payment level. pePFIt
is determined by amortizing the unpaid
principal. MCPFt _ h over its remaining tenn
and indexing the result to changes in prices.
Pit since the last adjustment:
(8)
PCPF, = CPF X MCPF,_,
X [1- (1
CPF)t-1'-l] -l
X [1
(Pt -P1-1) / P H ] ,

PCPF, =

t = I • . .. ,T-l
MCPF,_, (1
RVRM,).

t = T.

The unpaid principal is calculated by employing the same contract rate used to de-

•

termine interest charges with the standard
variable-rate mortgage:

(9) MCPF, = MCPF. IT (1
i :::: 1

+ RVRM,)

+ it~: PCPF,
X[._L (1 + RVRM.)]}.
-{ PCPF I

t = 1•...• T.
When CPF is less than RVRM. payments
start low and the principal is paid off at
a slower rate in the early term of the loan
than with either of the alternative mortgage designs. It is even possible for MCPF
to increase over part of the early tenn because interest charges may exceed payment credits.

Reserve Requirements-

Structure an Impediment
To Monetary Control?
Reserve requirements are widely
inherent in the present structure
viewed as the fulcrum of monetary of reserve requirements is analyzed
policy-the policy instrument that within the money multiplier hameenables the Federal Reserve Syswork. Next, recent management
tem to manage the nation's money of monetary policy is described,
stock by controlling the volume of noting how the Federal Reserve
reserves available to the banking
offsets changes in the relationsystem. Proceeding from this
ship between reserves and the
money stock in the course of stapremise, critics of the present
structure of reserve requirements
bilizing money market conditions.
have pointed to characteristics
Finally, a statistical analysis is
they suggest could impede, rather presented that indicates the curthan enhance, monetary controll
rent structure of reserve requireIn particular, they have noted
ments does not contribute importhat differences in the requiretantly to instability in the money
stock under current operating
ments on different sizes of banks
procedures.
and on different types of deposits can complicate monetary conStructure of requirements
trol and cause instability in the
The view that the structure of
money stock.
reserve requirements can disrupt
It is true that the relationship
monetary control is usually stated
between total reserves and the
in tenns of a money multiplier
money stock varies more under
present requirements than it would relationship in which the stock of
if requirements applied unifonnly money, M, is the product of the
to all deposits included in the
money multiplier, m, and the
money stock. However, whether the monetary base, B: 2
variability in the relationship is a
(1)
M= rnB.
significant source of imprecision
The monetary base is composed
of currency held by banks and the
for monetary control depends, in
part, on how the Federal Reserve public plus member bank deposits
exercises control. When the Federal at Federal Reserve banks. MonReserve offsets changes in bank
etary authorities can increase or
reserves to stabilize interest rates, decrease the base by buying or
selling U.S. Government secuthe structure of reserve requirerities-a process known as open
ments is not a significant source
of monetary instability.
market operations. When the FedIn this article, the potential
eral Reserve buys Government
problem for monetary control
securities, for example, it usually

RESERVE REQUIREMENTS
ON MEMBER BANK DEPOSITS

De~OIl1lnla ",al

(Million dollars)

Net demand deposits'

Less than $2 ..
$2to$10 ....
$10 to $100 ..
$100 to $400
Over $400
Time deposits'
Savings deposits ....
Other time deposits
Maturing in 30 to
179 days
Less than $5
Over $5 .........
Maturing In 180 days
to 4 years .........
Maturing in 4 years
or more . .......... .

Requ ire·
mant
(Pe rca nt 01
deposn s)

''''%

12
13

16'"
3

3
6
2Yz

Daman<:! depesitl ,ubja ct 10 r. se",a requirement,
a ra IIrou damln<l de poa lls minul cash Ilam l In
process 01 colla ctlon and d emand bal . nca. due
from doma. tlc b.... k•. ReQul<ement sche-dules .re
Ortdualad. and a.ch deposit Inle",. , app lies 10
Ih' l p,n 01 tha depo ~ lt l 01 e a ch bank.
2. The a .... 'a ge of reserve. on savl no s and other
tlma depO l ltl mull be al le llt 3 parcent. the
minimum I Pacllie-d by I. w.
SOURCE: Fad.,,1 Ru afV' Bull'titl
j.

pays for the securities by crediting the reserve account of a member bank. This increases the
bank's reserves-and, hence, the
monetary base-by the amount of
the purchase.
The quantity of money that can
be supported by a given base
depends on the value of the multiplier (m), which, in tum, depends

1. In one study. William Poole and Charles Liebennan suggest five changes in reserve requirements (along with four other
reform measures) that would improve the stability of the relationship between the money stock and reserves. See their
comprehensive study "Improving Monetary Control," Brookings Papers on Economic Activity, 1972. no. 2, pp. 293·335.
Numerous less technical articles also relate reserve requirements to monetary control. See, for example, William Burke,
"PrUner on Reserve ReqUirements," Business Review, Federal Reserve Bank of San Francisco. Winter 1974, and
J. A. Cacy, "Reserve Requirements and Monetary Control," Monthly Review, Federal Reserve Bank of Kansas City.
May 1976.
2. It is assumed here that the monetary aggregate to be controlled is the narrowly defined concept of the money stock.
M " composed principally oC private nonbank demand deposits plus currency in the hands of the public. A similar analygis
could be applied to other aggregates.
Business Review I December 1976

on a number of things. One is the
public's demand for currency.
Another is the member banks'
demand for reserves. The public's
demand for currency determines
the amount of the base that circulates outside banks. And the
demand for reserves influences the
amount of demand deposits banks
are willing to provide.
Within the multiplier framework, these factors are measured
in terms of four ratios: k, the ratio
of the public's currency to the
demand deposit component of
the money stock; r, the ratio of
required reserves at member banks
to these demand deposits; e, the
ratio of excess reserves at member
banks to these deposits; and v, the
ratio of vault cash at nonmember
banks to these same deposits.
The multiplier can be expressed as
a function of the ratios so that the
relationship above can be written: '

(2)

l+k
M= r+e+v+kB.

The argument that the current
structure of reserve requirements
hampers monetary control focuses
on the required reserve ratio, r,
and the characteristics of the present structure that lend instability
to this ratio:
• Reserve requirements at member banks vary with the level of
net demand deposits at the
individual bank. Five different
reserve requirements, ranging
from 7;.2 percent to 16;.2 percent, apply over five different
deposit intervals.
• Reserve requirements apply to
deposits other than demand
deposits. Requirements apply to
such diverse member bank liabilities as time and savings
deposits, Eurodollar borrowings,
and U.S. Government deposits.
• Reserve requirements for nonmember banks vary even more
than for member banks as they
derive from the laws of the 50

individual states. Both the
reserve percentages and the
types of assets that can be
counted as reserves vary from
state to state.
Most nonmember banks hold
correspondent balances with member banks. And the level of these
corresponcient balances depends,
in part, on the volume of demand
deposits at the nonmember banks.
Demand deposits at the nonmember banks indirectly give rise
to required reserves through their
effect on correspondent balances
that are subject to the System's
reserve requirements. Therefore,
the effect on required reserves of
a shift in deposits to nonmember
banks would be similar to that of
a shift to member banks with low
reserve requirements.

The argument that the current
structure of reserve requirements hampers monetary control focuses on the required
reserve ratio, f, and the characteristics of the present
structure that lend instability
to this ratio.
How these characteristics affect
the r ratio can be illustrated by
considering a check drawn on a
large member bank subject to the
16*-percent marginal reserve
requirement. Suppose the check
is deposited to a demand deposit
account at a bank with less than
$2 million in net demand deposits
and, therefore, subject to the
7t,2-percent minimum requirement. When reserve requirements
are calculated on the new deposits,
required reserves, R, fall by 9 percent of the amount of the check
while the demand deposit component of the money stock, D,
remains constant. So, the required
reserve ratio, RID, declines.

3. A simple derivation of equation (2) is in the accompanying technical appendix.

Suppose, on the other band, the
check is deposited to a savings
account, subject to 3-percent
reserve requirements. Required
reserves fall by 13 Y.z percent of
the amount of the check, while
demand deposits fall by the entire
amount. The required reserve
ratio rises.
Finally, suppose the check is
for federal taxes and the proceeds
are credited to the U.S. Government's tax and loan account at
the same bank on which the check
is drawn. Required reserves remain
unchanged because the same
reserve requirements apply to the
Government demand deposits as
to the account on which the check
is drawn. But Government demand
deposits are not part of the money
stock, so D falls and r (the
required reserve ratio) increases.
The reserve ratio does vary
under the present structure of
reserve requirements. But because
other changes can offset fluctuations in r, this variation does not
necessarily imply corresponding
variations in the money stock. For
example, if excess reserves were
to increase the same amount that
required reserves decreased, the
two changes would exactly offset each other and the multiplier
would remain constant. More
important, however, the Federal
Reserve, in the nonnal course of
its open market operations, usually offsets changes in the required
reserve ratio caused by shifts in
the composition of deposits.
Effect of interest rate control
The current operating strategy of
monetary policy focuses on shortrun control over money market
conditions-particularly the Federal
funds rate. The basic thrust of
monetary policy is quantified in
tenos of targets for monetary
growth, which, in tum, are shaped
in the light of the outlook for such

History and rationale of reserve requirements
Reserve requirements originated long before
monetary policy. In the early 1800's, state
legislatures began requiring banks to hold
coin in their vaults equal to a specified per·
centage of their outstanding bank notes. At
that time, bank notes accounted for a larger
portion of bank liabilities than did demand
deposits and, so, represented a more important claim on bank assets.
By the time of the Civil War, deposits
had overtaken bank notes, and the National
Banking Act of 1863 established reserve
requirements against both deposits and
notes. The Federal Reserve Act of 1913
retained the basic structure of reserve requirements of the National Banking System and incorporated some important improvements. For example, the act prohibited
member banks from holding reserves with
other commercial bank&-a practice that, on
occasion, had added to the severity of banking panics.
Reserve requirements were a fixed, constant percentage during their entire early
history. Then, amendment of the Federal
Reserve Act in 1933 authorized changes in
requirements under emergency conditions.
Further amendment in 1935 permanently
established variable reserve requirements as
a policy instrument.
In keeping with other moves toward centralization of authority at that time, the
1935 amendment stipulated that authority
over reserve requirements would reside exclusively with the Board of Governors of the
Federal Reserve System. Other policy instruments.-open market operations and the
discount rate-remained joint responsibilities
of the Board of Governors and Federal Reserve bank officials.
The use of variable reserve requirements
as a policy instrument was widely hailed as
a significant innovation and was subsequently adopted by several other countries.
However, with a few important exceptions,
changes in reserve requirements have been
infrequent and have played a minor role in
monetary policy.
The rationale of reserve requirements
during most of their history was to assure
banks a source of funds to redeem liabilities

Business Review I December 1976

that are payable on demand. At present,
almost 40 percent of all bank deposits are
demand deposits and, in practice, banks
also pay most time deposits on demand, In
contrast to these extremely liquid liabilities,
most bank assets are relatively illiquid,
Loans make up about 60 percent and securities less than 25 percent of bank assets,
and a portion of the securities are tied up
as collateral for government deposits,
Based on this difference in the liquidity
of bank assets and liabilities, the liquidity
argument for reserve requirements asserts
that banks must hold a portion of their
deposits as reserves in order to maintain
sufficient liquidity to meet deposit withdrawals, It was this rationale that prompted the
first reserve requirements in state laws.
However, reserves that banks hold to meet
legal reserve requirements do not contribute
significantly to the liquidity they need to
meet withdrawals, A member bank cannot
use such reserves to meet heavy deposit
withdrawals because legal reserves may not
be drawn below required levels. Except for
a 2-percent deficiency that the bank is allowed to carry over to the succeeding week,
any deficiency in reserves must be made up
within the current reporting week, This
means that if a member bank with a 10percent reserve requirement suddenly
loses deposits, only 10 percent of the withdrawal is freed through a reduction in required reserves.
Under current structure, even this descrip.
tion overstates the role of reserve requirements in providing liquidity. In September
1968, the Federal Reserve System intro·
duced lagged reserve accounting, specifying
that member banks hold reserves against
deposit levels two weeks earlier. There is
no longer any immediate tie between de·
posits and reserve requirements, and a proportionate volume of reserves is not freed
when a deposit withdrawal occurs. Under
lagged requirements, the decrease in required reserves resulting from a deposit
withdrawal occurs two weeks later.
The related view that reserve requirements assure the solvency of banks is also
debatable. Liquidity refers to the adequacy

of immediately available or readily marketable assets, whereas solvency refers to the
adequacy of all assets, liquid or otherwise,
to cover all liabilities, whatever their maturity. Since banks are so highly leveragedbecause they rely so heavily on borrowed
funcls-a relatively small decrease in loan
value represents a large change relative to
a bank's capital. Even a moderate decrease
in the value of the bank's loans can eliminate equity and produce insolvency.
While bank solvency is an appropriate
concern of the monetary authorities, reserve requirements are probably not the
most efficient means to that end. Capital
requirements, for example, are simpler and
more direct since solvency is measured

ultimate goals as price stability,
full employment, and satisfactory
economic growth.
Still, day-to-day implementation of policy focuses on the Federal funds rate. Each month, the
Federal Open Market Committee
establishes a target range for the
Federal funds rate that it estimates
is consistent with the targeted rate
of growth for money. The Trading Desk then generally maintains
the Federal funds rate within
that range.
To the extent that it constrains
this interest rate within a narrow
range, the Federal Reserve relinquishes short-run discretionary
control over reserves and the
monetary base. To keep the Funds
rate from rising, the Trading Desk
supplies additional reserves by
buying Government securities. To
keep the Funds rate from falling,
the Desk withdraws reserves by

against capital, not reserves. A more important consideration, however, is that reserve
requirements may not contribute to bank
solvency over a longer period. To the extent
that idle reserves decrease the income of
member banks, reserve requirements do not
encourage the accumulation of capital from
earnings or provide incentive for further
investment.
With wider agreement on these issues, attention bas focused increasingly on the contribution of reserve requirements to monetary control. That issue has acquired even
more importance the past few years as the
Federal Reserve has shifted its focus of
attention and begun to direct more effort at
controlling the money stock.

selling Government securities. Consequently, in order to control the
Federal funds rate, the Trading
Desk increases or decreases the
volume of reserves in response to
market demand.
Until the Federal Open Market
Committee began publishing
numericalr.angesforsho~run

operating targets in January 1974,
it was difficult to quantify the relative importance of the many possible target variables, though the
high priority attended. the Funds
rate was well known. Since that
time, records from the FaMe
meetings have included infonnation on numerical targets. As an
example, the relevant section from
the record of policy actions for the
meeting in May 1976 reads:
The members agreed that growth
in M, and M . over the May-June
period at annual rates within

ranges of 4 to 71h per cent and 5
to 9 per cent, respectively, would

be acceptable. They decided that,
in assessing the behavior of the
aggregates, approximately equal
weight should be given to M,
and M ..
The members agreed that until

the next meeting the weeklyaverage Federal funds rate might
be expected to vary in a gradual
and orderly way within a range of
5 to 5% per cent.·
As the accompanying chart
shows, since July 1974, the average value of the Federal funds rate
has fallen within the target range
specified by the FOMC every
month. 6 By contrast, the growth
rates for Ml and M2 , which are
averaged over two-month periods,
have fallen within their target
ranges only a third of the time
since July 1974. 'The monetary
base has never been formally targeted, but until recently, targets
have been set for a major component of the b~reserves available to support private deposits,

4. Federal Reserve Bulletin, July 1976, pp. 587·88
5. In April, May, and June 1974, the Federal funds rate traded at levels above the target ranges specified in the FOMC
meetings. In unusual moves, the FOMC raised the upper limit of the Federal Funds rate range in thjg period. The
Committee's willingness to allow the Federal funds rate to rise. instead of injecting reserves. enabled it to hit the RPD
target range in four of the six target periods in the first half of 1974. Over that span, both M, and M . were within
their target ranges in ea<:h or the six target periods. This evidenoo is too limited to draw a firm conclusion, but it does
suggest that large <:hanges in the Federal funds rate and close <:ontrol or reserves might pennit tighter control over
growth in the money stock.
12

or RPD's. The actual growth rate
for RPD's fell within the target
range only fivc times bctween July
1974 and March 1976, when the
targets were no longer included
in the policy directive.
This record indicates that, in
the short run, the Federal Reserve
has tended to alter the supply of
reserves in response to changes
in the quantity demanded. If
such changes had not been offset,
the Federal funds rate would have
been pushed outside the target
range as affected banks bid in the
Federal funds market to eliminate
undesired deficiencies or excesses
in their reserves.

Rang es of Interm e d i ate Targets and Ac hi ev e d Va lues

15 PERCENT--------------------------------------------

FEDERAL FUNDS

963 ~------------_.------------_.------------_.-

24 PERCENT-------------------------------------------'8
RPD·S

An open market operation that
offsets a change in required
reserves, so as to leave
money market conditions
unchanged, produces a
change in the monetary base
that, to a large degree, neutralizes the movement in the
money multiplier.

6 -

0-----------~~~__i4._f~----------------

-8~------------~------------~~---------~-15 PERCENT--------------------------------------------

For example, if the demand for
reserves happened to increase
because of a shift of demand
deposits to larger banks with
higher marginal reserve requirements, the desired borrowing in
the Federal funds market would be
greater than the lending. The
banks with reserve deficiencies
would then bid up the Federal
funds rate in competition for the
available reserves. The Federal
funds market provides an efficient
mechanism for redistributing
reserves among banks, but it cannot be a source of new reserves
to the banking system as a whole.
So, the initial result of excess borrowing pressure can only be a bidding up of the Federal funds rate.
AP, the Federal funds rate is
increased, banks might then tum
to other sources of reservesthrough selling securities or comBusiness Review I December 1976

'0 5 -

o--------~~~------~-¥----------------5

20 PERCENT-------------------------------------------

'5
'0
5
o~------------,-----~---__,---~~---,_

1974

1975

1976

SOURCES: Board 01 Governors, Federel Rue,.e S"tem
Fede,el Re,erve aenk of Oella$

peting harder for reserves by rais·
ing rates offered on time deposits.
As they sell securities to the public,
the reduction in demand deposits
would constitute a reduction in
the money supply.
But if the Federal Reserve tends
to offset the reserve deficiency
through an open market operation
that stabilizes the Federal funds
rate, and money market conditions
generally, there is no longer an
incentive for banks to sell securities to the public and no significant net change in the money
stock. An open market operation
that offsets a change in required
reserves, so as to leave money market conditions unchanged, pro·
duces a change in the monetary
base that, to a large degree, neutralizes the movement in the
money multiplier.'
Empirical estimates of offset
If changes in required reserves
caused by compositional shifts
in deposits under the current
structure of reserve requirements
are offset dollar for dollar by
changes in the monetary base,
there will not be a significant disturbance in the money stock. The
policy record verifying the TTading Desk's success in controlling
fluctuations in the Federal funds
rate shows that the Federal

Reserve has tended to alter the
supply of reserves in response to
changes in the quantity demanded,
thereby tending to accomplish
such an offset, in fact. In addition,
statistical regression analysis confirms that the degree of offset
provided by the nonnal operating
procedures of the Trading Desk
is very close indeed. 1
This analysis shows that since
January 1974, the Trading Desk
has, on average, responded to a
weekly change of one dollar in
required reserves by creating a
change, in the same direction, of
97 cents in the monetary base
through purchases or sales of securities. Furthermore, the difference
between the estimated value of 97
cents and the theoretically desirable one-for-one response is not
statistically significant. So, the
evidence supports the view that
the Trading Desk successfully offsets week-to-week changes in
required reserves.
The fact that changes in
required reserves tend to be accommodated on a weekly basis does
not imply that the Federal Reserve
loses control of monetary aggregates over longer periods. By
adjusting the tolerable range of
the Federal funds rate month to
month, the Federal Open Market
Committee uses interest rates to

influence monetary growth over
the long nul.
Conclusion
When the Federal funds rate is
contained within a narrow band on
a weekly average basis, as it has
been the past two years, different
reserve requirements on demand
deposits and requirements on
other bank liabilities probably do
not seriously disrupt monetary
control. Banks adjust to net
changes in required reserves in
ways that create pressure on the
Federal funds rate and induce the
Trading Desk to absorb or supply
reserves through open market
operations to maintain the rate
within its target range. As this
happens, changes in the money
multiplier caused by shifts in the
composition of deposits are largely
offset by changes in the monetary base.
Uniform reserve requirements
on all deposits that are included
in the money stock could permit
more precise monetary control if
that were desired.* However, Wlder
present objectives of open market
operations, reserve requirements
do not function as the fulcrum of
monetary control. Reserves and
reserve requirements do not fix the
volume of demand deposits. In
fact, the opposite is true. The vol-

6. The nature of the offset in terms of the multiplier model is described in the technical appendiJ:.
7. Details of this analysis are given in the technical appendiJ:.
8. If policy were conducted so as to achieve precise short-run control over the monetary base instead of the Federal funds
rate, changes in the structure of reserve requirements that make the relationship between reserves and the money stock
more stable would then make a difference for monetary control. A more stable and predictable money multiplier
would lead to a more predictable money stock, given the base, and. hence, facilitate monetary control.
Of course, precise control of the money stock has not been the sole short-run objective of the Federal Reserve. When
the money stock departs from its targeted path, the Federal Reserve generally attempts to bring it back on course only
gradually because the alternative would requi:re an adjustment in the Federal funch rate deemed too large and disruptive
to be consistent with the maintenance of orderly financial markets.
Nevertheless, if the current practice of smoothing short-run fiuctuatiolUl in interest rates were abandoned in favor of
more precise control of the money stock over short periods, recent studies suggest that-given the current structure of
reserve requirement&-an operating strategy focusing on the monetary base would be about as effective as one keyed to
the Federal funds rate. See, for eJ:ample, William R. McDonough, "Monetary Policy-Effectiveness of Alternative
Approaches to Monetary Control," Business Review, Federal Reserve Bank of Dallas. August 1976. Therdore, steps to
reform the structure of reserve requirements so 88 to make the required reserve ratio more stable than at present might
tip the advantage to a reserve aggregate strategy and allow more accurate short-run monetary control than is
currently feasible.

ume of deposits depends on interest rates and other variables.
Reserve requirements determine
the volume of reserves the Trading Desk supplies or absorbs to
maintain interest rates in the target range.
Just as reserves depend on the
level of deposits, the base depends
on the total money stock. Some
observers believe that increases in
the monetary base are a leading
indicator of increases in the money
supply. This argument would
have some merit if the base were
independently determined. But
when reserves are supplied in

response to current demand, the
base is determined by the money
stock-rather than the other way
around. The base then becomes a
coincident indicator of currency
and, because of lagged reserve
requirements, a lagging indicator
of demand deposits. Any value it
has as a leading indicator is
accidental.
-Edward E. Veazey

Technical appendix

The basic multiplier model expresses the
money stock (M) as a multiple (m) of the
base (B):
M=mB.
The expression for m in terms of ratios
can be derived by substituting the elements
of the money stock and the base in the
money multiplier relationsrup. The money
stock is composed of demand deposits (D)
plus currency held by the public (C). The
base is composed of required reserves at
member banks (R) plus excess reserves at
member banks (E) plus vault cash at nonmember banks (V) plus currency held by
the public (C). Substituting terms and dividing by D, the multiplier is obtained as:

M
(D +C)/D
m = F = (R + E + V + C) / D'
Division by D makes it possible to express
the multiplier in terms of four ratios-r =
RID. e = EID. u = VID. and k = CID.
Thus:
l + k

m = :-,,+-Ce;;-+T-:u;-+"'
k'

and

l+k
M = , +e+ u+kB.

Busine8IJ Review I December 1976

The impact of a change in required reserves can be analyzed by differentiating
the multiplier model with respect to R:
dM
dmB
dB m

dR =dR

+ dR

Consider a change in required reserves
that is caused by a shift that leaves total
demand deposits unchanged. (This would
be the case, for example, if demand deposits
simply shifted between banks of different
sizes.) The currency ratio, k, would not be
affected. If the demand for excess reserves
and for vault cash depends on such factors
as interest rates, deposit variability, and
total demand deposits-rather than on the
distribution of deposits among banks-the e
and v ratios would also remain constant. In
this case, dm/dR can be evaluated as:
1
k
1
m
dm

dR = -(r + e + u+ k) ' 1)=-)3'

and substitution then gives:

dM _-m+dBm.
dR dR
If B were exogenous (if dB/dR = 0), the
disruptive impact of an increase in R would
be to reduce M by m times the change in R.
However, when the base is adjusted by the

Trading Desk to offset changes in R (that
is, when dB / dR = 1), then the two components exactly offset one another, and

dMldR=O.
It is possible to test the Trading Desk's
response to changes in required reserves by
estimating a model of open market operations. The equation estimated is:
t::.G = bo bll:J.R
b2l:J.C
bsl:J.O
u.
The variables are defined as follows:
l:J.G = change in Reserve bank holdings of
U.S. Government securities
l:J.R = change in required reserves
l:J.C = change in currency in circulation
flO = change in the composite of other
variables that also affect bank re.
servesl
u = random disturbance, assumed to
have an expected value of zero
The equation was estimated with a least
squares regression on first differences of
average weekJy levels, beginning in January
1974 (when numerical ranges for interest
rates and monetary aggregates were first
announced) and ending with June 1976. To
minimize the possibility of simultaneous
equation bias stemming from the possible
endogeneity of the composite variable, 0,
we created 0·, an instrumental variable
that excludes the effect of Wednesday data.
That is, in place of the daily average of 0
for the entire statement week ending
Wednesday, we use 0·. which depends on
the daily values only through Tuesday.
By Wednesday, the Trading Desk has
fairly accurate data for the major factors
affecting reserves for the first six days of
the statement week and bases Wednesday's
open market operations on those data. Using O· in place of 0 in the regression reduces

the possibility that the value of the regressor is detennined simultaneously with open
market operations. The value of O· has
already been determined when the Trading
Desk determines the level of l:J.G for the
statement week in a final adjustment in
Wednesday's operations.
The required reserve variable, l:J.R, presents no problem since it is predetermined
under lagged reserve accounting. The currency variable, l:J.C, is assumed to be determined by factors, such as seasonal demand
for currency, that are not affected by the
level of open market operations. Also, the
Federal Reserve can measure changes in C
by simply monitoring its own shipments of
currency. Therefore, it is assumed that AC
is exogenous and its value is known.
We are interested in determining whether
the structure of reserve requirements has
contributed to the instability of the money
stock. There is little doubt that variation
in required reserves caused by deposit shifts
would complicate monetary control if the
Trading Desk were attempting to maintain
the monetary base on a predetermined path.
However, if each change in required reserves
is matched by an equal change in the base,
the two changes would tend to offset one
another. The variation in required reserves
would not then be a significant source of
monetary instability.
Since open market operations have a direct impact on the reserve component of the
base, we can test the hypothesis that the
Trading Desk offsets changes in required
reserves by testing the hypothesis that the
coefficient of IlR is equal to 1. The regression results favor acceptance, rather than
rejection, of that hypothesis. The regression
results are given in the accompanying table.

1. Weekly average holdings of U.S. Government securities, currency in circulation. and all variables
eJ:cept net reserve carryover that are included in the composite 0 are in the Federal Reserve Bulletin
table "Member Bank Reserves, Federal Reserve Bank Credit, and Related Items." Carryover is reported
in the Federal Reserve release H.4.1. Required reserves are in the Bulletin table "Reserves and
Borrowings of Member Banks."
The composite variable is calculated as: 0 = float + other Federal Reserve assets + acceptances
and discrepancy (calculated by subtracting the itemized components of Reserve bank credit outstanding
from the total) + gold stock + Special Drawing Rights certificate account + Treasury currency
outstanding - Treasury cash holdings - deposits, other than member bank reserves with Federal Reserve
banks - other Federal Reserve liabilities and capital + member bank currency and coin reserves + net
reserve carryover.

MULTIPLE REGRESSION ANALYSIS OF OPEN MARKET OPERATIONS
(Dependant variab le _ aG)

Ordinary least squares estimate

Con.tanl

8.53

(39.52)
Estimated vari ance of
coefficient, assuming
random coefficient model
Ordinary least squares estimate .. .

.,.
-68.15
(473.48 )

.97

(.OS)

1.04
(.14)

.05

.0'

1.11
(.10)

1.04
(.17)

Ind..,endent varlabl..
W

-26.40
(48.36)

30.99
(35.771

-1.09
(.32)
.08
-1.10
(.04)

NOTE : Figural In parentl"ue, &11 'iindard e rror. oflhe coefflcllnts.

The estimate of bb the coefficient of all, is
.97-with a standard error of .08.
Thus, the empirical evidence is consistent
with the hypothesis that when a shift in
deposits between banks with different requirements gives rise to a deficiency or
surplus of reserves, the Federal Reserve accommodates the shift by supplying or withdrawing reserves in open market operations. Similarly, when fluctuations in time
deposits, U.S. Government deposits, or other
bank liabilities create changes in required
reserves, the Federal Reserve largely accommodates the new demand. The structure of reserve requirements influences the
volwne of open market operations, but, on
balance, there is no significant effect on the
money stock.
A similar argument applies to an increase
in currency demand by the public, given
its demand for the total money stock. When
demand for currency increases and depositors increase their currency holdings by
drawing down their checking accounts, the
transaction-in addition to changing a bank's
required reserves-also changes the bank's
available reserves. ~ But if the Federal Reserve exactly matches the flow of currency
from its vaults with open market transactions (in addition to offsetting the change
in required reserves), the transaction probably does not destabilize the money stock
to a significant degree.
Although currency in circulation, C, includes vault cash of nonmember banks, it
consists mainly of currency in the hands

of the public. Therefore, we can test the
response of the Trading Desk to changes
in the public's demand for currency with the
statistics already estimated. The hypothesis that the coefficient of t::.C in the regression equation equals 1 cannot be rejected.
Our estimate of this coefficient is 1.04, with
a standard error of .14. These results are
consistent with the hypothesis that the
Trading Desk offsets changes in C that
might otherwise be the cause of instability
of the money stock.
An assumption underlying this regression
model is the standard one that the coefficients are constant. But when this assumption is applied to the behavior of the Trading Desk, it assumes away an interesting
question: Is the response of the Trading
Desk to changes in reserve requirements
uniform over all observations? If the demand for reserves varied unpredictably with
changes in required reserves, the response
of the Trading Desk would have to vary also
in order to stabilize the Federal funds rate.
Under these circumstances, it might be
argued that the structure of reserve requirements added to the difficulty of controlling
the Federal funds rate.
To test the stability of the Trading Desk's
response, we altered the basic assumption
of the regression model to allow for the
possibility that the coefficients vary over
time. In this random coefficient model, each
coefficient can be written:

b, = b*,+ d"
where b· i is the expected value of the co-

2. Lagged reserve accounting changes the timing, but not the direction or amount, of these changes.

Business R eview I December 1975

efficient of the ith explanatory variable and
d. is a random component with assumed
zero mean and constant variance. Thus, for
all observations, it is assumed that:
E(d.) = 0 for all i
E(t4dl ) = 0'" for i = j
= 0 otherwise.
Unbiased and consistent estimates of the
variance of the coefficients were obtained
with a method due to Clifford Hildreth and
James P. Houck.' The estimated variances
of the required reserve coefficient and the
currency coefficient are .05 and .04, indicating relatively stable responses.
Evidence presented in the text suggests
that the Trading Desk has been much more
successful in hitting its interest rate target
than in maintaining the monetary aggregates within their target ranges. The Federal funds rate has fallen within the target
range every month since July 1974, while
the monetary aggregates, Ml and M 2 , have
fallen within their target ranges only about
a third of the time. This evidence supports
the hypothesis that the Trading Desk allows
smaller fluctuations in the Federal funds
rate relative to the target than in the monetary aggregates.
It is possible to examine this hypothesis
more formally by adding other explanatory
variables to the model of Trading Desk operations and testing for the relevance of
other influences on open market operations.
Of particular interest is the response of the

Trading Desk to deviations of the monetary
aggregates from the targets specified by the
Federal Open Market Committee. The
Trading Desk continuously monitors the
growth rate of money and might be expected
to adjust open market operations when
the current growth rates deviated widely
from target.
We reestimated our model of open market
operations to include two additional variables, T 1 and T 2, which measure deviations
of the growth rates of the monetary aggregates from target levels. When these variables are included in the regression model
of open market operations, neither is significant at a lO-percent level. The hypothesis
that these variables do not significantly influence the volume of day-to-clay open market operations cannot be rejected.
Of course, this does not imply that monetary growth rates do not infiuence policy.
The FOMC continuously monitors monetary
growth rates and considers past growth and
estimated future growth before deciding on
the appropriate monthly range for the Fed~
era! funds rate and other target variables.
Thus, the range that the FOMC adopts
for the Federal funds rate is based
partly on desired long-run monetary growth
rates. However, in day-to-day open market
operations, there is no significant response
in the amount of reserves provided to
the deviations from two-month monetary
growth targets.

3. "Some Estimators for a Linear Model with Random Coefficients," Journal of the American Statistical
Association 63, no. 322 (June 1968)

New member bank
Hays County National Bank, San Marcos, Texas, a newly organized institution
located in the territory served by the San Antonio Branch of the Federal Reserve
Bank of Dallas, opened for business November 5, 1976, as a member of the
Federal Reserve System. The new member bank opened with capital of $400,000,
surplus of $400,000, and undivided profits of $200,000. The officers are: William C.
Carson, Chainnan of the Board; John N . Cunningham, President; David M.
Edwards, Vice President and Cashier; and Steve Hadlock, Vice President.

New par banks
Caldwell Bank & Trust Company, Columbia, Louisiana, an insured nonmember
bank located in the territory served by the Head Office of the Federal Reserve
Bank of Dallas, began remitting at par November 1, 1976. The officers are:
E. L . Carroll, Sr., Chainnan of the Board; R. J . Lee, Executive Vice President;
R. E. Chapman, Vice President and Cashier; Ronnie Darden, Vice President; and
Margie Fore, Vice President and Assistant Cashier.
Texas Bank of Amarillo, Amarillo, Texas, a newly organized insured nonmember
bank located in the territory served by the Head Office of the Federal Reserve
Bank of Dallas, opened for business November 15, 1976, remitting at par. The
officers are: J. Richard Hankins, President; Rhea F. Raines, Vice President and
Cashier; and Cecil Mason, Vice President.

Burneu Review I December 1976

Federal Reserve Bank of Dallas
December 1976

Eleventh District Business Highlights
DEMAND DEPOSITS
TURN OVER FASTER

Rapid inflation and high interest
rates the past few years have
resulted in more sophisticated asset
management by the public. Sound
economic management dictates
that idle funds be reduced to a minimum and that most funds be held
as earning assets.
This is evident in the management orthe public's cash balances
at commercial banks. Demand
depoeit.8 have become more expensive for people to hold in recent
years, as inflation cut deep into the
purchasing power of money and
interest rates rose sharply.
The rapid in6atiah of the past
four years has increased. the
nominal cost of household and business operations. Since most payments for goods and services are by
check, use of demand deposits at
banks has risen substantially. Consequently, the dollar volume of
checks written has risen sharplymuch more than the average level
of deposits. Stated another way, the
annual rate of turnover of these
accounts-the number of times each
dollar of demand deposits is spent
during a year-has increased
sharply.
Each month commercial banks in
the 30 largest metropolitan areas in
the Eleventh District report data
on demand deposit accounts of individuals, partnerships, and corporations and of state and local governments. These groups hold the bulk
of demand deposit accounts.
In 1970, depositors used demand
deposit balances an average of 38
times. By the end of 1972, however,
inflationary pressures began to
build rapidly. Prices and interest
rates escalated, and speculative
buying by consumers and business-

men mounted to a fever pitch. The
increase in spending pushed debits
up sharply. At the same time, the
growth in demand deposits slowed
markedly. And by 1974, the annual
turnover rate had risen to nearly 55
at District banks.
In 1975, the economy moved from
reression to recovery. Short-term
interest rates fell sharply. The
three-month Treasury bill rate, for
example, feU to 5.8 percent from 7.8
percent the prior year. Moreover,
the rate of inflation declined to
almost 9 percent from about 11 percent, and real disposable income
inched upward.
As a result, demand deposits at
District banks began to expand at
a faster pace, after slowing two
consecutive years. The growth rate
of debits, however, slowed sharply,
mainly because business-still
keenly aware of excessive spending
in the two previous years-assumed
a cautious stance toward further
spending. Therefore, the turnover
of demand deposits rose only
slightly from a year earlier.

DEMAND DEPOSIT TURNOVER
64 ANNUAL RATE OF TURNOVER-

48-

BUSINESS LOAN DEMAND SOFT

32 -

_'7o....llL,~7~2---:'74

1976 based on flrat tan montha on',

With a continued slowdown in
inflation from very high levels,
most short-term interest rates
declined about 50 basis points more
in the first ten months of 1976, and
total demand deposits in the District picked up. As the recovery
progressed, spending by both
businesses and consumers rose
moderately. The ratio of debits to
demand deposits resumed climbing
steeply, 80 the annual turnover
reached 61.
The turnover of demand deposits
has risen considerably faster in the
largest metropolitan areas in Texas.
In the first ten months of 1976.
customers of banks in the Dallas
area, for example, turned over
their demand deposit balances at an
annual rate of 89 times; in the
Houston area, the turnover rate
was 69. Historically, these two areas
have accounted for well over half
the total debits to demand deposits
in the 30 reporting centers in the
District.
Since Dallas and Houston are
the major business and financial
centers in the District, it is not surprising that debits to demand
deposits are higher in these areas.
Large metropolitan areas have a
high concentration of businesses.
and buainess accounts generally
have a faster turnover than others.

.,.-

With few exceptions, demand for
business loans at weekly reporting
commercial banks in the Eleventh
District h as remained weak in 1976.
Many industries have made sizable
net repayments of their bank
indebtedness, and total 10808 to
businesses increased only 1.6 per.
cent in the first ten months of the
year, compared with a decrease of
(Continued on back page)

INDUSTRIAL PRODUCTION
(SEASONAllY ADJUSTED )

TOTAL PRODUCTION -

140 (1967=100 )

140 ( 1967=100) - - - - - - - -

./"

"\ TEXA~/.:

131.t

130

...I t30.4

-V. . . ..

120

-MINING-

MANUFACTURING-

130

:, h ' ., .:

........
-'
...... ' I

130.0

\
111.8

.....

U.S.

'.
'

v. ··· · . ./\

100 \ LOUISIANA

LOUISIANA '.;

90 -

110 -

T EXAS

.•••.••.

'~":

110

115.4

........

110 _

~.."
U.S.

~
!-

.,
't>"..1..~

\ ( ..... . . ·. \~.t'~ \V
120- ¥

120 (1967=100)

137.0

96.6

./\.!

\ ...

LOUISIANA

100 .---:-::-:-:--,---:-::=-o--,-...,..,-=--,1974

1975

80-,-~~.-~~~~~

1976

1974

1975

1976

SOURCES: Board of Governors, Fed eral Re serv e System
Fed eral Re serve Bank of Dallas

EMPLOYMENT AND UNEMPLOYMENT

PRICES RECEIVED BY TEXAS FARMERS

FI\I£ SOUTHWESTERN STATES'
(SEASONALLY ADJUSTED , BY FRB)

270 (1967=100)

9.0 MILLION - - - - - - - - - - - - PERCENT 8
UNEMPLOYMENT RATE

t:/·\.. · ·.. ····~ ~~:~HT SCALE) 8.8

8.9-

8.8-

-6

8.7-

...

180
150

LIVESTOC K AND
LIVESTOCK PRODUCTS

1975

1974

-5

1976

SOURCE: U.S.Oepartm.nt of Ag,lcl,lltu,e

...

-4

8.4-

8 .3

210

120

8.68 .5 -

240

-,---:-::c::-:--r---:=,----,--=,.,---,
1974
1975
1976

1. Ar izona, Louisi.n •• New Me~ico, Okl ahoma. and Texas
SOURCE: State employment .ge ncles

SAVINGS AND LOAN ASSOCIATION ACTIVITY
AND HOME BUILDING IN TEXAS
(SEASONALLY ADJUSTED. BY FRB )

900 MILLION DOLLARS - - - - - - - PERCENT 100

700 -

190 ( 1 9 6 7 = 1 0 0 ) , - - - - - - - - - - - - -

180

150

....

;« ....... .

140
130

" . . " I •I' I'
,

80
70

_.... "''''173.3

400 -

-./'\.... --S- AVINGS (LEFT SCALE)

50
40

···i69.0

/.~.~.:................ ~ -. HOUSTON
..
U.S.
'

....... DALLAS

-,---:;:,,-;--.-----:=:--,--:-::c=--,1974
1975
1976

SOURCE: U.s. 81,1r881,1 01 L.bor SI.tistic.

500 -

_. . . ~ . . =:.~.--. . ... .

160

/\1

WITHDRAWAlS·TO ·
SAVINGS RATIO
l
\ .. (RIGHT SCALE) 'I

182.0

...

170

600 -

I,~ ,""
~

800 -

CONSUMER PRICES

300 -rI-~1=9=74~--,-~719~7~5~--~--719~7~6~--r
12.0THOUSAND--------------------~---

10.08 .0 •.0

7 .•

4.0
2.0

SOURCES: Bur • • u 01 8uline • • R... . rch. Univ.,.lly 0' T.u.
Feder.' Hom. lo.n S.nk 01 Little Rock

CONDITION STATISTICS OF ALL MEMBER BANKS

RESERVE POSITION OF MEMBER BANKS

ELEVENTH FEDERAL RESERVE DISTRICT
(CUMULATIVE CHANGES)

ELEVENTH FEDERAL RESERVE DISTRICT
(MONTHLY AVERAGES OF WEEKLY DATA)

3.0 BILLION-DOLLAR CHANGE - - - - - - - - -

250 MILLION DOLLARS - - - - - - - - - - -

2.5-

200 LOANS

2.0-

1976

1.5-

1974..,.······: :

-././~~;:·~·C~-·~"r-'"-"·r·-"-·~r·:"·:"·;' .'·"~;:'~O-~'C·~"'~r·~=' =';~'C·~_'r>_'_;r·_'7_5~

BORROWINGS
FROM FRB

A :.

.~N\J\
1\7E\~: ~:\ET
.
'''''''''''.'~: : _'1
, i~

:~:

-5~ -

:. .

¥-

-1002 .5 BILLION-DOLLAR CHANGE - - - - - - - - INVESTMENTS

2 .0 1.5 1.0 _

1976

___: _ _

-,---;-;=-,---:::;;-;--,--=,,---,
1974
1975
1976

.- -_ ..- -

~>.~4
'''''=''''''''''
~ ~"~
,,

.5 -

-150

197;....
--

~ \../~./

3.0 BILLlON·DOLLAR C H A N G E - - - - - - - - -

LOANS AT WEEKLY REPORTING BANKS
ELEVENTH FEDERAL RESERVE DISTRICT
(CUMULATIVE CHANGES)

2.5 TIME DEPOSITS

2.01.5 -

1974 .'

..-;;..-::
1976
, .....
·CCc,' .....;;"....
............... ..

-':;'':;~''''- ::-::.:--';15-

1.0 -

800 MILLlON· DOLLAR CHANGE - - - - - - - -

400 -

.5 -

O~~-r,-~~r-~~~~~
2.0 BILLION-DOLLAR CHANGE - - - - - - -1.5 -

.....
DEMAND DEPOSITS

1.0 .5 -

.....

-.5 -

1976
J

I,
F
M

I
M

I
A

I
I
I
SON

...--

1975

200 MILLION-DOLLAR CHANGE - - - - - - - - CONSUMER LOANS

~"';'974'"

o - _••• ;;;.
• ""'._
••a"...
..,__
,.,":"..~:;_:_2-~-~---'.-7-5--~-.::-~-,,- 100

1974
I
A

, ___ -..

- 200-~r=T-'--r~-'-r-~~~-'--r-

~L:::
.'
~.;.::=ri"h""·

- 1.0

~ ___ ...

100-

1975 _ .... "
...... -

/'"

197~••..•·'··"·
_/' ;~';,"="

BUSINESS LOANS

- - -,r-",r--,r--,r--,
r--,r--,r--,r-C,!-,-O",'
-"0 J
F
M
A
M
J
J
A
SON

I
0

FOREIGN TRADE
HOUSTON CUSTOM S REGION
(SEASONALLY AOJUSTED, BY FR8 )

BUILDING CONTRACTS
FIVE SOUTHWESTERN STATESl
(SC ASONALl Y ADJUSTED. BY FRB )

1.4 BILLION D O L L A R S - - - - - - - - - - - -

1.2 BILLION DOLLARS - - - - -- - - --

.g.
.s - "',..

1.2 1.0 -

.s .n

.6 -

,.....

'\1

I \ I
\"

, . .' .... ' .... , i\ I

",
•

IMPORTS" "

.4-'
o
.2
1. A.izona, louls'ana, New Mu ico, Ok lah oma, and Te xa s
SOURCE: F. W. Dodg a, McGr aw-H ili, Inc .

-,---;;;-;:;----,-=;;-,-----;=
-,1974
1975
1976

SOURCE, U.S. Oepartment of Comm arc e

0.5 percent in the corresponding
period last year.
The slow growth in demand for
business loans this year probably
reflects both an increase in the supply of funds available from other
sources and a decrease in the need
for bank funds. Increased sales and
profit margins during the recovery
have resulted in sharply higher
internal cash flows. Moreover, some
companies have used the commercial paper market or have refunded
short-term debt with borrowing in
bond markets.
A reduced need for short-term
funds has depressed the demand for
bank loans. Business inventories
have been reduced, and many firms
used a portion of sales to repay
bank debt. Inventory levels are
being closely monitored to prevent
involuntary buildup in inventories.
The only major sources of
strength in business loan demand in
in 1976 have been the oil and gas
industry and retail trade. The rate
of increase in mining loans slackened from the extremely rapid pace
a year ago, as drilling activity
slumped through last May from the
peak level at the end of 1975. But
with a rebound in drilling, the
petroleum industry has continued
to make heavy demand for bank
funds to finance exploration and
development of oil and gas fields.
Demand for bank loans by
retailers has trended upward all
year as merchants cautiously raised
the level of their inventories. An
acceleration in loan demand was
evident in the fall, when retail
inventories were expanded in anticipation of an improved Christmas
buying season.
Loan demand by manufacturing
industries as a group was flat in the
first ten months of 1976. Output of
manufactured goods, as measured
by the Texas industrial production
index, declined sharply in the
second quarter and did not recover
to the peak established last March
until the end of the third quarter.
Overall, producers of durable goods
have made a small net repayment
of their bank debt, despite sizable
increases in borrowing by the primary metal and transportation
equipment industries.
Loans to manufacturers of nondurable goods have risen only

slightly this year. A sharp increase
in loans to producers of textiles,
apparel, and leather and a moderate gain in loans to manufacturers
of chemicals and rubber were
almost offset by declines in loans
to producers of other nondurable
goods.
Transportation, communication,
and public utility industries have
reduced their bank debt considerably this year. The decline in loans
to public utilities has been especially sharp and reflected, in part,
the continued refinancing of shortterm debt in the capital market.
Bank loans to the construction
industry declined 12 percent
through October this year, even
though residential building, especially of single-family houses, has
been strong in the District. However, a pickup in the demand for
funds has become apparent as
building construction has begun to
show more strength.
OTHER HIGHLIGHTS:

• The Texas industrial production
index, seasonally adjusted,
remained virtually unchanged in
October from a month earlier,
according to preliminary data.
Manufacturing continued to
recover, while mining output was
down sharply.
In durable goods manufacturing,
substantial output gains were evident for all major industries except
stone, clay, and glass. In the case of
nondurable goods, strong advances
were recorded by the apparel and
printing and publishing industries. Output of petroleum refining
and paper products industries
weakened.
Crude petroleum production
accounted for the decrease in mining output. Drilling, however, increased for the fifth month in a row.
• The unemployment rate for the
five southwestern states declined to
6.0 percent of the total civilian
labor force in October from 6.3 percent a month earlier. The decline
was the second in a row and placed
the jobless rate at the lowest level
since March.
Total employment increased for
the third month in a row, reaching
the highest level since January
1975. The statistics suggest that
the rise can be traced largely to em-

ployment of unemployed workers
already in the labor force, since
unemployment has fallen sharply
for two consecutive months.
• The value of total construction
contracts in the five southwestern
states reached an all-time high in
October. The increase centered in
nonbuilding construction, largely
because of contracts for two electric
power and heating plants in Texas
and Louisiana that totaled $875
million.
The total value of contracts for
structures in October was at the
highest level since April 1975.
Nonresidential building contracts
climbed nearly three-fifths from
September. Residential building
contracts were more than 6 percent
above a month before.
Housing starts in Texas fell to
7,800 units, seasonally adjusted, in
October. The decline, however, followed the unusually high level of
starts in September that resulted,
in part, from increased Government-financed housing activities.
• Total loans and investments at
member banks in the Eleventh District rose in October for the fifth
consecutive month, as bank loans
expanded at the fastest rate for any
month since November 1973. The
gain largely reflected a sharp
increase in demand for real estate
loans and continued strength in
loan demand by consumers and the
oil and gas industry.
Member banks slightly increased
holdings of U.S. Government obligations and moderately reduced
holdings of other securities in
October. In addition, the banks
stepped up borrowings in the
Federal funds market to finance the
increase in loan demand.
The growth of deposits at District banks slowed slightly in
October. Two factors have contributed to the slowdown in deposit
inflows. First, interest rates offered
on time and savings accounts at a
significant number of banks have
been drifting down since last June.
Moreover, many banks have
switched their advertising campaigns promoting time and savings
deposits to encouraging loans.

1976
JANUARY

FEBRUARY
MARCH

Bicentennial PerspectiveDecline and Fall of the Gold Standard
Home Building in TeIa~
Analyeis of Housi ng Decline Suggests Prospects of a Good Recovery

District BankingNew Edge Offices Participate in Exp8JIding International Banking Market
Review of 1975
Treasury Cash BalancesNew Policy Prompts Increased Defemive Operations by Federal Reserve

APRIL

MAY

Bicentennial PerspectiveDevelopment of Capital Markets in the United States
Meat Production_

Grain Price Increase Accentuates Beef and Pork Cycles
Bank LiquidityIs the Level Adequate for Future Loan Expansion?

JUNE

Fiscal PolicyCrowding Out Estimated from Large Econometric Model

JULY

Bicentennial PerspectiveDevelopment of the Texas Oil I ndustry
Industrial ProductionNew Louisiana Indez Dominated by Mining

AUCUST

SEPTEMBER

OcrODER

Monetary PolicyEfCectiveness of Alternative Approaches to Monetary Control
Electric PowerUtilities Look to Increased Use of Coal and Nuclear Energy
The Payments MechanismA Primer on Electronic Funds Transfer
Urban DevelopmentA New Approach
Interest RatesU.s. Government Securities Reflect No Increngc in Uncertainty

NOVEl'tIBER

Grain SuppliesIssue of a Reserve Stock Program tor the United States
Discount WindowSeasonal Borrowing Privilege Liberalized

DECEMBER

Rea] Estate FinanceAdvantages of Innovations in Variable-Rate Mortgages
Reserve RequirementsStructure an Impediment to Monetary Control?

Business Review
Federal Reserve Bank of Dallas

Full text of Review (Federal Reserve Bank of Dallas) : December 1976

FRASER