View original document

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

Working

Paper

76-1

BANK ~UillAGEMENT, COMPETITION, AND INTEREST RATES:
A PORTFOLIO MODEL OF DISCOUNT \vINDOW ACTIVITY

William

Federal

Reserve
July

Jackson

Bank of Richmond
1976

The views expressed here are solely
those of
the author and do not necessarily
reflect
the
views of the Federal Reserve Bank of Richmond.

Introduction*

The ability
valuable

to borrow

incentive

for

behavior

Fifth

they appear

at,

that

extensively

borrow

they differ?

the

maximizing
extends

should

financial
paper

institutional

changes.

The resulting
using

time

and

borrowing

determine

why

organizations

If

so,

how do

criteria

activity?

important

The answers

of bank portfolio

behavior

to
in

of receD,t years.

Capital

Asset

portfolio

model of a bank as a utility-

This

model of bank risk-return

Pricing

demand for

borrowing

do not?

and microeconomic

a theoretical

varying

to

Do banking

that

our knolwledge

investor.

markets,

window.

over

recent

organizations

member bank borrowing

develops

the

a potentially

Reserve System.

markedly

paper examines

macroeconomic

window is

Federal

varies

from those

climate

financial

empirically,

credit

discount

enrich

the well-known

discount

to the

banking

the

individual

questions

This

District

differ

Are both

stormy

of
This

or avoid,

in determining
these

source

organizations.

of large

at the

a bank to belong

Member bank use of this
among banking

funds

Model to encompass imperfect

financial

services,

and interest

model of bank behavior

at the

discount

management

is

rate

tested

window as the

dependent

variable.

Portfolio

Choice

of

The Capital
competitive,

especially

Competitively

Traded

Asset

Model explains

unregulated

Pricing
markets.

This

Securities

starting

portfolio

choice

point

the

for

*The author is indebted
to Emily Cart,
Peggy Nuckols,
Marsha Shuler for computational
assistance.

in

analysis

and

of

2.
2.

bank portfolios
for

proofs

is

of this

based on several

assumptions

(see [18,24,31,55]

model).

They are:
(1)

Investors

choose portfolios

single-period

time horizon,

Their

of holding

utility

random variables:
standard
(2)

Investors

interest

rate

assets

to

asset

only

(3)

utility
a(R):

risk

(5)

transactions

Interest

This
equity

valuation

a

based on the

and risk--expected

asset

that

at a low
risky

dominate

risk-

sets.
efficient

portfolios

with

a(R).

positively

class

with

within

which

their

E(R) and negatively

with

marketable,

with

aversion.
are divisible,

and taxf1~ee.

estimates
rates

horizon:

basic

opportunity

choose a risk

identical

(6)

is

a riskless

of assets

a given

costs,

with

within

framework.

can be combined with

choose Markowitz

Securities

planning

or lend

This asset:

investment

varies

assets

1:'eturn E(R),

form combinations

Investors

operate

a (R) .1

RF.

maximum E(R) for
(4)

financial

can borrow

Investors

they

uti1ity-maximizing

expected

deviation

as if

Investors

are not E!xpected to
security

returns
wide

analysis.

no

are price-takers

of 1~eturn and risk

model has found
and portfolio

salable

for

any asset.

change over

the

are not state-dependent.

acceptance
Its

in the

implications

theory
may be

1The tildes
denote random va1:,iables.
Differentiation
with
respect to random variables,
which assume values according
to a
prescribed
probability
density function,
is of course impossible.
Investors
define
utility
over the anticipated
mean and standard
deviation
of such distributions.

~

of

i

3.

clarified

by examining

behavior

is

the behavior
to determine

aggregated

Under

of individual

market relationships.

the conditions

postulated

by assumptions

individual

investors

of competitive
(1)

follows

through

investors,

security

(6),

the

the continuous,

whose

markets

utility

twice

function
differentiable

form:
K
k
-U = f
[E(R),

where the

utility

increases

with

of the k-th
expected

investor

return,

au

but

a(R)],

from holding
decreases

>0.

au

"ai"(R)'
Rational
functions.
requires

As the
a larger

probability
aversion
isoutility
space.

investors

is

defined
locus

is

to

An individual

bears

compensate
investor's

by his

locus

convex

to the risk

utility

of constant
axis

more risk,
for

he

the higher

,

degree

of risk

utility.

This

in risk-return

That is,

where second order

conditions

> 0

2

,

a
a

-

E~!~
a(R)2

> 0

are met by the second partial

term..
Conservative

they will

<0.

risk-averse

investor

return

a E(R)
a a(R)

derivative

possess

representative

of loss.

risk:

-a-am-

thus

expected

with

a portfolio

accept

only

investors
low risk.

prefer

low expected

Aggressive

investors

returns

since

seek high

ofI

4.

expected

returns

assumptions
strive
for

they

maximum expected

with

market

relationship

return

asset,

emerges.

tradeoff

of risk.

A is

Rr is

the riskless

return
be in

a positive

the

for

exceeds
excess

them again

suffer
hold
loci,

falling
identical
their

the rate

of

that

predicted

on the
is

less

than

prices.

_

rise

appropriate
all

between

utility

[~=A'
.~)J

is

price

Assets

the slope

optimum tradeoff

market

portfolios

market

line.

'I

is:

zero,

market

until

for

portfolios
Similarly,

their

investors
of all

risk

and

thus

line

should
containing
assets

class

will

do not have to

investors'

expected
is

of risk,"

whose expected

capital

market line.

Although

portfolios,

change of

"the

prices

"capital

and its

+ A 0' (R),

by the

capital

shows the

e:fficient

of interest.

demand; their

lie

whose return

rate

of risky

risk-return

on an asset

defines

constant,

or minimum

holdings

reuationship

--

investors

risk,

by combining

E (R) = ~

where

given

a det:erminate

line

Following

risk-averse

the return

The tradeoff
market

high risk.

all
for

This

between

The capital

if

return

the riskless

equilibrium

accept

however,

maximum expected

assets

degree

which,

(2) and (3),

for

risk

for

return

isoutility
and risk

when

5.

With

returns,

constant

expected

the

investors

absolute

utility-maximizing

can

assume

an

risk

aversion

and

normally

portfolio

exponential

form

relationship

such

where

the

k-th

parameter

investor's

by

his

own

risk

AMB.

toward

this

asset

return
low

all

low

points

by

Line

CML

it

line

Chart

is

1

>

is

originates

such

as

as

at~.

O.

tangent

to

the

portfolio

expects

investor

along

utility

investors

maximize

high

They

proportion
crc ).

of

They

investor

lie

a

this

lend

asset

portfolio

closer

to

asset
to

to

assume

2

See

functions
taking
generate

[24].

implies
falls

when

negative

higher

The
unrealistic
wealth
marginal

than

average

common

risk.

Their

of
along

investors

assumption
behavior.

increases
utility.

of
It

and

that

high

very

utility
that

high

returns

CML
who

expected

quadratic
suggests

the

with

proportion

riskless

holds

~.

utility

low

the

this

who

when

their

hold

down

obtainable

return

that

segment

M denotes

receive

loci

set

moves

highest

increases

isoutility

portfolios.

(R

to

the

c,
prefer

The

determined

The

It

An

return
a

6).

appears

combination.

made

and

(p.

Point

Conservative

assets

clarified

combination.

cases,

axis.

risky

(R)/2)],

c

market

portfolio

risk

are

risk,

capital

risk-reward

"average"

investments

the

-ccr

avoidance:

combinations

until

portfolio"

In

at

axis

risk

"market

risk

only

Market

risk

efficient

be

asset

Capital

the

[-c(E(R)

investor

along

may

risky

The

-exp

preference.

theory

efficient

-

U(R)]=

denotes

position

This
of

c

for

as:2

--2
E[

distributed

risk

fJ)
104

fJ)
Q)

0
~

> .~
~u
.~
0

.-i

Q»,

=

,~

"""'

t)

-0

t-I

~

+

-<

Z

~

~

<H
~

fJ).-i

>~
orojoroj

H~
~

fJ)oroj
t-I~
~~

104~
CCO

fJ)
~

~

oroj

Q).~

~CO
~~
US
orojO

'~.o

fJ)

oroj<l;O

bO

co

.-i

~ ~oroj

104 fJ) Q)

0 >,
oY.oY.

~

Q)~

~

C/)

~
t-I
~

k2
b

-

'

6.

"0

...~.-i

co

b

0
~

Q)

.-i

~

0

u

.
u

.c

104
<U

~

.-i

H

.c

uco
OJ

Po

oroj

.-i
co
~

<

~

OJ

~

~

104

oroj

t-I ~

0.

>,104
oY.O
fJ)

U

~8:

I

I
I

I
I
~

~

~ ~

U

bO

~~
Q)Q)

oroj

fJ)

>-.04
~-

oroj

0

>104
<UfJ)

oroj.-i

~Q)oroj
0>
U
U ~ 0

fJ)
Q)~ fJ)

104

=

~C/)
~
Q)~

~

fJ»,

0

t-I

H

~.o
<U

.c

I
I
I

S

I
I
I
~

0

>.~

-~

U

~

III

-~~~
~

N

U
~

bOfJ)
COoroj
-Q)

«

Q)~

II""

co

I

U
~

~

oroj
104

OoY.
~
fJ)
fJ)oroj

OJ 104

>
104

104

OJ

~"O

c""\oroj

.c

I\UOJ

"""'

,~
~

<U

104

N<U""
~Q)Q)
1\1o4P.-i
0

~

"

C)

=

U~

~
co

~
oroj

<U

~
>0
H

t-I
~

~

~

-k2

Q
~
HZ

U~,
~ ~
~H
~~

--

7.

return
find

portfolios
that

are supported

their

utility

are expanded

by high

these

along

points
This

certainty

portfolios

such as (Ra,cra).

equivalents

the behavior

markets.

decisions,

investors

to points

describes

in bond and equity
and operating

leverage

Aggressive

.along CML when their

is maximized

CML are

theory

by borrowing.

It

can also

including

real

In particular,

it

portfolio

of regulated

of individual
analyze

financial

behavior

Banks as Institutional
Unlike
marketable
basic

securities.
is

supplying

information

obtained

securities.
than

less

that

perceived

securities)

Banks issue
at rates
no explicit
charge

almost

about

~

to

budgeting

and

to analyze

riskless

a bank.

(2),

the

their
along

a

superior

and "customer

banks issue
than

(deposits)

In particular,

in

high

return

such loans

investors.

i~1 less

securitie~1

time

lending

]~isk of extending

by individual

Their

(5),

in nonmarketable

perceived

and traditionally

lying

to assumption
in

only

price-takers;

bank loans

to invest

whose return

invest

entirely

experienc:e

assumption

or less.

interest

income to

banks
Their

And contrary
(secondary

Contrary

through

allows

is

capital

financial

institutions.

bankf; do not

indivisible

demand curve.

primary

business

can be modified

Banks are not

downs loping

relationships"

investors

Investors

most individuals,

business

of

of ~.

sector

merger proposals.

All

the

liabilities
lending

bearing

rate.

interest

most demand deposits
represent

clnd savings

pay

a source

of service

deposits

represent

8.

a riskless

(Federally

insured

in

most cases)

pooling

of funds,

asset

to conservative

nonbank investors.
The resulting
information,

specialization,

and lending

is

and transactions

responsible

for

mediaries.

Correspondingly,

investnents

available

securities-only

could.

and equity

To the

Market

extent

elastic
Line.

adjusted

return.

It

quantity

setting

firm

an efficient
This
set

preferred
receive
not

that

Line

necessarily

alternative

Its

from realizing

would

earns

set

invest

in

investors

underlie

it

face

the

same

the Capital
an excess

setting

an unregulated

risk-

and partly

of 'this
full

If

level

of risk,

return

would

remains
model,

potential

a

hold

Line

[17].

opportunity

an unregulated
it

would

be (RA- ~).

intermediation,

bank

expect
It

to
would

since

the

available.

regulation

return.

Market

investment

intermediary.

investment

bank could

the Capital

anyone by this

context
their

it

the higher

excess

"exploit"

of direct
In the

also

private

a rate

dominate~;

to assume the market
RA'

loan!~,

partly

reflects

to a financial

return

that

that

thus

opportunity

bank could

markets

9) shows that

portfolio

available

the

[50].

2 (p.

Bank Market

risky

do so,

extends

is

of

would

financial

But as it

Chart

it

inter-

bank dominates

securitiel;

that

locus

investment

An unregulated

debt

infinitely

efficient

in borrowing

of financial

opportunity

to an unregulated

[35,40,49,50,55].

costs

the existence
the

Markowitz

any of the

based on divisibility,

may prevent

Reserve

banks

requirements

/ '

~

,<

0

/

~

, /\

/

/~
H

~<I'

~

<
~~

~

~E-t

E-tH

...Z
0

r..1
z

~
/

/

/

/

\

0/

/

r..1

u:
H

tJ)

~

u
E-tr..1

H~
...::J

r..1E-t
ZH

~
~ tJ)

~
~

:z ~
H~

I
\.

~

~

.
9.

~

9.

~<U

'"
v,

~~
<Uq)
s.. S

~<U

<U

OCCU

t: fJ)
~O

t:.o

~~
<U.

fJ)

s..
.t:~b

."

~

<U

0.0:':

~
fJ)

~~

<0

s..5

~<U

t:

>
II-j

fJ)<U

O.t:

fJ)~fJ)

CU

t:

OJtUfJ)
t:.t:<U

~fJ).-I
='OJfJ)
O~o..t

. S

s..

~

fJ)~

~
t:

~

s..fJ)<U

~

<U t: fJ)
.t:o..to..t

~

tU

'"
t:

H

o..t

~
o..t

>

='

.-I

H

~

~

='

6'Z

~
~

~

\

~

~
~ fJ)
fJ) s..
>'<u

/

S

~s..<u

b

t:t:

.t:

Oo..ts..
CU
~

t:

.-I

>'6

UtU~

bOt:<U

.-IfJ)<U
=,~1I-j

~~

~

QltUs..

s..

0

t: fJ) s..
OfJ)<U

U::JCU

<uS~

s..o..t<U

fJ)OC~

~o..tQ

~

tU

o..t

.-I1I-j

o..t

~1I-jt:

c.;

~

t:
<U

~~tU

s..oo
~
s..

~.o

CU
CU

.t:

U

.os..~

0
P-

.00.0

I

~

I

~~
CU

<UtUU

~S
~s..O

I

='.t:fJ)

&~~

I

.-I<U<U

I

.I

I

I I
I I

I

I

~~':y;

< CU
~

I
I

~

.cd"

~<

N

t:o..tt:
CU~o..t

N~

o<uo

<U
o..t~
U~

~
t:
CU

\

~

t:

I

s..

I

0

~

~
Z

/

/0

/ ~~ /
~Z

<0

Q
~

tj~
~::J
P-E-t

rJ~

10.

clearly

lower

asset

beyond

return

expected
that

returns,

required

on a portfolio

since

for

transa(~tions

containing

i and E(Xi)

r

on portfolio
falls

the proportion

denotes

the expected

cash is

clearly

proportionally

terms,

holding

instead

with

required

of secondary

lower

certainty

Wi

less

reserves

proportion

to total
market

risk

to risk,
investor

along

the Capital

restraint

represent

in

The zero return

expected

cBLshholdings.

asset

return

In risk-adjusted

in th,e form of primary
portfolio

reserve

profitability

assets
to a

the

capital

of portfolio
risk

beyond

2.

size,
from

could

If

in

reserves

a regulated
RA to ~

be less

same risk

than

the Proportional
are held

bank's

[49].
that

preference

requirements,
assets,

in

reward

If

the

for

bank is very

of a conservative

whose portfolio

further.

certainty

prohibitions

and other

some absolute

speculative

earnings
the

Chart

appears

lies

Market Line.

on earning

risk-adjusted
Line

return

with

Moreover,

in

declines

its

private

assuming

Hence,

reserves

portfolio

from

invested

from asset i.

than~.

reduces

of required
Line

types

return

reserves

Bank Market

certain

The weighted

E(i i )'"

of the portfolio

portfolio

Reserves

averse

purposes.

equivalent.

The effect

assuming

an inefficient

n
=

i=l
represents

always

n a~;sets is:

~
E(R)

where wi

cash is

(All
equivalent

regulations

level

returns

[27].

does not

points

against

lying

of the

holding

prevent

banks

The resulting
necessarily
along

zero-risk

reduce

a Bank Market
intercept

rate.)

-

11.

11.

These restrictions
strengthen

public

application
tends

limit

accounting

confidence

of risk-asset

to reinforce

risk-averse

solvency

are not

as though

is

to

not

traded

utility

preference."

goals

[17,

p.

judgments

of banker

prudence,

isoutility

loci

Bank utility

functions

involve

positive

offset

utility

the dollar

liabilities

The risk

will
full
true

sheets,

of

liability

33,

and Risk

36,

Since

"bank management behaves
and attempts

to

about stockholder

regulation,

with

and stockholder

the properties

marginal

shown above.

utility

of return,

cost,

and negative

risk.

Increased

returns

costs

of assuming

marginal

are required

to

increased

38,47,51].

Aversion

preference
holding

functions.

liability

and psychological

[2,22,

Bank Portfolios

balance

or

and

764.]

generates

of asset

utility

markets,3

of its

They try

profitability,

det:ermined

preferences

utility

profit-maximizers.

of liquidity,

in "perfect"

on the basis

marginal

Regulatory
in particular,

have homogenous! expectations

The combination

negative

ratios,

they

by banks.

unconstrained

subjectively

stockholders

maximize

capital

behavior

conflicting

according

bank stock

while

Investors

Banks thus
the

of return,

the b.anking system.

and other

Banks as Institutional

to reconcile

in

riates

of individual

interregional

banks may be suggested
and intertemporal

variations

by their
in

3"Simply attempting
to maximize the market value of the stock
not yield
the highest
utility
if the market does not reflect
the
value of the bank and if its stockholders
are unable to obtain
the
value by liquidating
their holdings."
[17, p. 764].

12.

demand constant
liability

Vault

Federal
e

10,12,28].

Consider

the following

asset

and

schedule:

Cash:

~

[2,9,

Cash,

Other

Borrowed

Funds Sold

Correspondent

Federal

Funds Purchased

Government

~

Other

Deposits

~

='
~

Government

~

'Pol
-~
~

='
co

~
4J

cx:

Securities

Certificates

of Deposit

~

CI)

~

~

~

4J

Securities

IPC Demand Deposits

'Pol

0

.Q

IPC Time and Savings

Deposits

Fixed

Assets

Capital:

Total

Assets

Total

Financial
This

simplified

of assets
return,

and operating
balance

sheet

and liabilities

curve

prevails

More liquid
corresponding
capitalized

of

to
value.

in

these

accounts

may shift

a bank.

characteristics

conditions.

assets

Illiquid,

receive

credit

risk

returns.

assets,

equipment

used in operations,

the

require

lower

and high

somewhat risky

portfolio

(The cost,
if

a downsloping

money periods.)

generally

low or zero

Fixed

for

the risk-return

"normal"

tight

cx:

Equity

are identical

orders

~

Liabilities=Assets

under

and volatility

yield

risks

Debt,

~

~
~

Loans

cx:

largely

loans

returns,
predictability
earn

the

highest

nonmarketable

j~ven higher

expected

of

plant

and

returns

than loans.
Conversely,
risk

~

~

'Pol

<

;:I

~

cu

~

~

~

~
~

Balances

Reserves

of withdrawal

more liquid
or

unavailability

assets

possess
generally

higher

volatility.

increases

as liability

The

13.

volatility
to

increases

short-run

from debt

withdrawal,

through

liabilities.

t-lhi1e the

require~nts)

may decrease

sloped

yield

available

curve,

l'Any increase

in

dollar

the

portfolio

paribus_,

deposit

should

find

both

accumulates
the risk
Hence,

or

may sell

in

drains

secondary

severe

be sought

maturity

connection

between

adjustments

only

rising.

enough to

or volatility
of operations

in

the

discount

While
assets

a bank's

of higher

longer

term

bank risk

preference

thus

occurs

through

the balance

short

run.
the
in

cash drains,

at the

by the promise

running

deficiencies

short-run

raises

returns.

banks

window,
may

and liabilities,
overall

risk

The

and shorter

term

sheet identity:

4
Kaufman's study [34] orders the variability
of deposits
as
follows:
U.S. Government deposits>
demand deposits
due to banks>
other demand deposits
~ other time deposits
~ total
demand deposits>
total
deposits>
time and savings deposits>
savings deposits.

-"

'--'-'.'"c.

'"

cc.c.

p.

thus

raises

create

of their

[49].

Such a bank

acceptance

borrow

of the

I' [33,

thereby

levels

management activities.

any category

and should

assets,

exposure.

funds,

risk

a positively

Ceteris

aggressively

risk

To meet these

reserve

liability

to match the

to lend long"

liability

reserves.

or engage in other

risk

and/or

reserve

deposit

set.

risk

based on volatile

for

management"

opportunity

overall

short-term

size

or aggregate

aggregate

subject

given

total

"liability

cash below required

liquidity

banks

reserve

loans
its

or operational

higher

and its

asset

legal

attempt

increases

return

its

to the

individual

a bank's

short

retains

increasing
of

worsens

not

liquid

increases,

at:tached

of

is

(ad.1usted

1jlquidity

increa~;ing

"borrows

of depleting

probability

with

volatility
that

oj: liabilities

as their

the volatility

unambiguously

A bank

cost

which

4 to highly

depol;its,

uncertainty

rises

balances

and eq1~ity capital,

120].

14.

r assets

= r liabilities

are attainable
although
in

only

longer

in

provides

by accepting

term

the long run.

assets

and equity.

low risk

earning
risk

in

be almost

the

ratios

short

run,

can be attracted

as important

[2,28,49,50].

on these

asset

and !'capital'!

should

profits

new evidence

liability

deposits

Liabilities

determining

Higher

as

(The Appendix

relationships.)

Demand Variations
These portfolio
along

a given

avoidance.

Bank Market

reserve-adj

us ted

Bank Market

Line

that

level

It

result

attain

environment

Line,

environment.

assets

differences
bank risk

should

also

vary

both

into

returns

available

somewhat riskier

a higher

level

ratios

of utility
rise

the barriers

bank structure.

[49].
to entry,

in

Chart

to all

assets

available

The resulting
returns

from

that

underlies

the

3.
then

or liabilities.

encourage

banks

They can clearly
profitability)

relaxation

customer

growth.

higher

banks

(risk-adjusted
This

High Demand

income or population

demand curve

Line

usted

in demand may result

each bank to achieve
bank loan

"normal"

returns

increase

from a concentrated
allows

the previous

efficient

This

High Demand Bank Market

encompasses

individual

and a reserve-adj

shows the greater

less-elastic

Loan/deposit

interbank

15) depicts

of economic activity:

The higher
to shift

3 (p.

Bank Market

from a high

the higher,

varying

bank earning

Chart

in a more favorable

may also

describe

and over time.

For example,

steeper

Line:

The demand for

cross-sectionally

market

preferences

[24,

of assumption

relationships,

and

(5)

36].

/

~~

~

Z
1-1
~

§ ~

/>o~

/'

1-1

.o~
~@

"';"--'

15.
15.

.

U)

~
>

~~ H
cet/)

~
~o;

~"V

'O~
~o;

~~
~~ ~
I

&g.

~
HIQ
~~
~ o.-!
~ 0
H

~

H
~p...

CUo

~p.

I-I~

~

'00

~>;
> ~

~~

~~

~~
HH

8
.0 ~

~
~

~

.90~

U) 0
Ho.-!
~~
~IQ

CUt/)

~
~

0.-1>;
~U)

~.3
1-1

IQCU

>0;
~
H ~

0

~~

U)S

H >;
bOU)
CC~
CU~

0
\
~

0

'
""""-.e-'

~

M'g
OCU

tSZ1

~U)

~
H~

~~

~

~

~

(J.o

\\

\

,
.

alS

~

o.-!~
U)O

.0

IQ

~

~
CI)

~

~~

~ ~ ""~ ~
<

0

""~;':::
~

~~

\\
0
\
'

t§~

\

\

~

~

~

~
~

~

>

~,..

\

'0
\

\

\

0

~;'
Z
~~

\0
\
..\
~

~
~

iii

~\

~
.<'

~

'",

~
1iIV-

~rv~
iii

v

~c:..

~~

iii

~

~~

>..c

~o.-!

0,..1

~

PI

E

IQ~

bO

.0 S
IQO"",

~~
~~~~

,

/~~
Z~

'-

,
,.

!j~

"/

~

/

~

~
~
~

;,:::~

I-I~
~~

/' t§~
~
/
,

~>;

~~
~ ~
CUO

.oS
~~

> ..c

o.-IbO
U)o.-!
U)~

gg

~
CU

~

.0

t~
~~
~~
~~

16.

restrictions

on branching

transfer
to

the

resources

across

realistic

related

to

that

economic

health

rate

on Bank
of

~furket

in
Market
it

increases

Line

[24].

This

financial

investors

The

can

required

then

of

lines.

It

profitability

geographic
climate

banks

is

highly

territory.
even

more

dramatic

money raises

the

risk1ess

the

and slope

the

Capital

rises

height

have

and rotates

>0.'

receive

to

conforms

When tight

line

~
a~
All

its

ability

as state

a bank's

of

Lines.

interest,

the

general

such

that

Variations
effects

the

distances

observation

the

limit

of

when

RF increases:

'i!-C!-ffi
>0.
aRF

a higher

return

at

the

same risk

level.

E (R) rises:

~~~-

>0

at equilibrium.

oRF
The Bank
to

assumption

rates
to

rise,

(6).
the

accommodate

particularly
becomes
relative

~Iarket
Bank

the

to

by extensive

security

sloped
returns

liability

their

portfolios

banks

may increase

loans

[44].

management

despite
their

stimulus,

[33].

Their

loan

rise,

Tight

money periods

behavior.

Banks

and industrial

rates

when banks

contrary
[51].

increases.

commercial

disintermediation
aggressive

this

state-dependent

usually

customers
when

to

highly

demand for

long-term

positively

strongly

are

bank

rising
of

reacts

returns

demand for

that
more

Line

as loan

seek
[38,44].
This

to

attempt
loans,

supply

returns
are

pattern

function
increase

also

tyPified

or

expand

maintain
Even

As

conservative
is

shown

in

17.

Chart

3 when the

steeply

sloped

riskless

Tight

rate

increases

Money Bank Market

to RC.
Line

The higher,

stimulates

more

bank risk

acceptance.
Bank Risk,

Utility,

and Borrowing

Member bank borrowing
a bank.

It

but

the

also

cost
of

involves

only

forms

Reserve

of

the

cost

opportunity

of other

Federal

not

represents

of

explicit

liability

management,

to borrow

earning

assets,

marginal

cost
Federal

Reserve

of the

is

subjective

represents

[30]

discount

the

and the

to
rate,

opportunity

implicit

cost

"
the marginal

pressure

element

minus

the

pretax

pecuniary

return

on

of borrowing.

a strong

such as loans

cost

of disuti1ity

asset management,

Bank "surveillance.

The incentive

a source

on extensively

to the

the amount of borrowing

true

cost

and psychological

borrowing

banks adds

of borrowing.

and C represents

If

the dollar

B
cost

of borrowing:

~ <0
dB
'
according
this

to [2,

reluctance

preference

21,47].

Only the

to borrow.

and interest

~ <0
ac'

rates

lure

Accordingly,
should

2

and ~
a

B2

of high

returns

both
influence

>0

individual
borrowing.

overcomes
bank risk

18.

The Financial

Climate

When the
profit-conscious

discount

rate

banks should

intermittent
rate

and Borrowin~

benefit

below earning

lags

behind

rising
to

be attracted

of Federal

Reserve

money market
discount

the

window.

System membership,

asset

rates,

rates,

RF in

One

a discount

then becomes available.

bill

rates,

(See Chart 4,

p.19).
As Treasury
climb

in

tight

opportunity
assets

money periods,

cost

of any bank along
tend

to exceed

distance

Federal

funds

its

position

exceeds

Federal

funds

incentive

the

rate

their

48,

rate
the

Bank Market

its

total

extensions

discount

tight

discount

of borrowing

Model,

rises.

The
reserve

3,

the

Line

position
would

vertical
line

utility

would.

by borrowing

then borrow

~

buy

of loans.

bank reserve
in

Chart

by a greater

bank would

of other

exceeds

rate

Pricing

secondary

Money Bank Market

increase

larger

In

a "normal"

cost

disutility

position

instruments

money periods.

When the

rate,

reserve-deficient

overcome

banks

by the resulting

profit

to borrow.
Empirical

level

the

its

bank loans

l3J.

discount

A more aggressive

to fund

usually

[7,11,

along

Asset

by selling

Tight

bank could

run.

Moreover,

may find

increases

the prevailing

than

short

reserves

the resulting

Even a conservative
in the

the demand for

of adjusting

simultaneously

the Capital

of interest
53J--and

with

studies

find

rates--financial
interest

rate

that

borrowing
demand [1,

may vary
2,

5, 11,

differentials--opportunity

with

both

13,15,22,

the
30,

P ercent

19.

$ Billions

13

4.0

11

3.0

Member bank borrowings
from Federal Reserve
Banks (right scale)

9

2.0

7

1.0
5

'I
3

3-month Treasury bill rate
on new issue (left scale)

0
Discount rate -FRB
Daily avg. rate (left

N. Y.
scale)

1

1967

1968

1969

1970

1971

1972

1973

1974

1975

Chart 4.
Adjusted aggregate borrowings, the
Discount Rate, and the Treasury bill
rate, January 1967,
through June 1975. (This borrowings series removes
the emergency credit extended to the Franklin
National Bank in 1974. The amount by which borrowings
by large New York City banks from May 1974, through
October 1974, exceeded the corresponding 1973 amounts
is subtracted from the aggregate series for a nonconfidential
adjustment.)

,

20.

costs

of

asset

and liability

profit-seeking
most of
money"

[19,

Treasury

is

is

that

more

loans.

the

The Supply

discount

with

of

Federal

[29]).

of

"...

periods

of tight

to

their

of greater
of

Treasuries

when rates

speculate

are

on capital

investment

when loan

gains
should

demand falls;

relationships

demand for
the

demand for

in

borrowings

discount

rate

borrowings

most markets

generally

in

Chart

relationship

varies

[4,42,
directly,

4.

Borrowings
of member banks when they

window appears

funds

of the

business

Conversely,

form of low' risk

derived

The accommodation

for

that

support

direction

use of funds

to price-quantity

not inversely,

tion

to

liquidation

incentive

most profitable

The aggregate

borrowed

form

then decline.

contrary

the

their

29].

This

The corresponding

53].

during

borrow

Indeed,

[3, 6,

Government securities.

borrowings

is

because

banks have a profit

represent

A occur

banks then

holdings,

to borrowing

falling,

by the observation

of Regulation

unlikely

toward

related

supported

This

p. 61].

security

utility

in

is

the violations

It

is

behavior

management [14,15,20].

at the existing

discount
Reserve

or monetary

credit--is
conditions.

their

term,
isoutility

an elastic

supply

rate

56].

discount

window--whether

Over the longer

banks decreases

to create

it

is

not supposed
(For

[39,
"open"

approach

schedule

for

The administra-

or "shut"

to vary

exceptions

"counseling"
loci

initially

with

to this

of extensively

away from the return

to requests

changes

in

policy,

see

borrowing
axis.

This

21.

"surveillance"

is

individual

essentially

bank borrowing

The Determinants

run.

the

result

of prolonged

[46].

of Borrowings

Borrowings
short

endogenous,

thus

are

primarily

The more aggressively

demand-determined

in the

a bank seeks profitable

loans,

the lower

will

be its

effective

liquidity,

and the more it

utilize

the

discount

window.

The higher

its

into

loans,

more it

relative

would

characterize

to its

be expected
individual

incoming

cash flow

to borrow.

outgoing

is

likely

cash flow

from deposits,

the

These relationships

banks and the banking

system [2,

3, 7, 9,

10,14,16,20,29,32].
"Aggressive-management"
items,

such as Federal

and time

deposits

positively
and internal
held,
savings

borrowings,

section
the

ratio

related
that

is

supported

analysis

of all

of borrowed
to 'volatile"

comprise

control.

connection

internal

or highly

certificates

of

deposit

Less volatile,

such as cash,

cash should

accounts,
reserves

to

balances

other

securities,
related

portfolio

items,

somewhat atheoretical

required

His

study

reserves

and is

negatively

or

subject

are

to

bank management.

member banks [41].
reserves

be

fund-providing,

correspondent

"attitudes,"

by Morgan's

loans,

should

be negatively

more "conservative"
between

volatile

Government deposits,

Government securities,

and vault

reflecting

This
borrowing

items,

funds sold,

deposits,

purchased,

to borrowings.

reserve

Federal

funds

including

related

fund-using

and

crossshows that

is

positively

related

to

to greater

items

bank

to

.~

22.

Empirical

Sources

of Borrowing

Borrowing
marizes

amounts

of borrowed
as the

behavior

and frequency

reserves

dependent

managerial

variable

for

The hypotheses

empirical

explanatory
microbanking

studies,

of 80-95% of the

is

initially

used

examined.

financial,

borrowings.

are of the general

to required

and the inevitable
"e'l tends

observed

ratio

and

The supply

elastic.S

to be tested

of borrowed

variables

are

the competitive,

BIRR = f (Xl'

where the ratio

figure

sum-

analysis.

of the demand for
be fairly

that

the averaged
This

variables

represent

determinants
should

reserves.

independent

largely

of borrowings

of borrowing:

to required

Twenty-eight
These variables

can be shown by one measure

x2'

...,

reserves
error

to be extremely

linear

form:

xn ),e~

depends
term "el'.
large,

on n
(In most

on the

order

variation.)

5
The possibility
that inter-District
differences
in borrowing
behavior
may reflect
varying
attitudes
of discount
administrators
has
been raised
[37].
Such variations
would amount to regional
monetary
policy.
It is more likely
that computed inter-District
differences
in borrowing
indicate
differences
in bank attitudes
toward the discount
window, bank structure,
or the demand for bank services.
These
microeconomic
factors
should be more important
determinants
of borrowing than the nonprice
rationing
that would occur if some discount
windows were open wider than others [2, 39, 52,52].
Moreover,
all of
Hinderliter's
[29] attempts
to quantify
non-price
rationing
in timeseries
analysis
were unsuccessful.

.-.23.
The portfolio
are portrayed

characteristics

by the variables

of these

available

from

banking
reserve

organizations
position

accooots:
ASSETS

CASH

Actual reserves plus
process of collection
deposits;

CaRR

Correspondent
balances
total
deposits;

DEMDEP

Demand deposits
deposits;

FFPUR

Federal funds
total
assets;

purchased

FFSOLD

Federal
assets;

sold

GOVDEP

u.S.
total

Government
deposits;

deposits

GOVSEC
I
'.

The average
organization;

asset

U.S.
total

Government
assets;

securities

LOANS

Loans and discooots

foods

size

of each banking

cash items in
divided
by total

held

divided

divided

by

by total

divided

divided

by

by total

divided

by

divided

divided

by

by total

assets;
OTHERSEC

Other securities
bonds") divided

(such as "mooicipal
by total
assets;

SAVDEP

Savings deposits
deposits;

TlMEDEP

Time deposits
(including
deposit)
divided
by total

USSEC

u.S. Government securities
divided
Government deposits:
a proxy for
Government securities;

VCASH

Vault

cash divided

divided

by total

certificates
deposits;

by total

of

by u.S.
unpledged

deposits.

24.

The popular
portfolio

image of large

managers

smaller

banks.

might

suggest

It

of securities

involves

transactions

[9, 16].

Sone. of these

Others,
with

lower

borrowing

41].

since

proportional

to

organizations

includes
banking
that

for

desired

in

that

generate

(zero-one)

organization:

mu1tibank

higher

J"

"",~,"",","

the

of funds

should

depress
may be
into

be less

output

returns

of micro-demand
variables

companies

that

exhibit

banks is

Virginia
holding

"""

of these

should

account,

than

banking

stimulate

borrowings.

effects,

the analysis

show the

location

tested

all

of each

The possibility

different

banking;
companies.

--

"""..,""",

should

[10].

[44].

branching

6They dominate
are multibank

Virginia

low cash

relationship

assets

factors

sources

banks,

DC, MD, NC, SC, and VA.

holding
large

smaller

funds,

borrowing

to nondeposit

the demand for

the existence

binary

than similar

size

economies

relatively

of banks are taken

reserve

is

and other

to stimulate

characteristics

deposit

Variations

than

more than

of alternative

The assets-borrowing

when other

particularly

costs

borrow

fluctuations,

such as the
tend

aggressive

and borrowing

preferences,

bank access

transactions
[16,

negative

To test

would

such as larger

deposit

risk

factors,

banks,

size

and availability

of cash management,

of larger

they would

between

relative

economies

ratios

that

The relationship

more complex.

banks as highly

borrowing

behavior

by the VA variable.

firms

in
.-.,

the

sample
'-?

6

from

~"'",.,

25.

Financial
rates

demand variations

and interest

rate

are represented

differentials.

by interest

They are:

DRATE

Fifth
District
discount
rate,
under
sections
13 and l3a of Regulation
A;

FFRATE

Federal

funds

rate;

FFLD

Federal

funds

less

TBILL

New issue

3~onth

Treasury

bill

rate;

TLD

New issue 3~onth
discount rates.

Treasury

bill

less

The higher

the level

be borrowing.
is

usually

in

the

Although
associated

discount

ferentials

should

Q2. Q3.

Banks

that

in

an increase
lessens

rate.

the

rates;

the higher

discount

rate

in borrowing,

should

itself

an increase

the FFLD and TLD interest

dif-

depress borrowing.
borrowing

48.53.54].

are examined

interest

an increase

with

rate

Finally.
[13.20,

of any single

discount

activity

may follow

The possible

by binary

variables

seasonal
that

seasonal
influences

represent

patterns
on borrowings

each quarter:

Ql.

and Q4.

Examined

The sample
organizations

consists

consolidated

of 23 large

companies

are large

management.

They hold

approximately

in each location:

~~""

District

on a member-bank basis.

bank holding

deposits

Fifth

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

enough to practice
three-quarters

the District

banking
These banks

and

liability
of member bank

of Columbia.

Maryland.

.'C"

North

"

Carolina.

c",

26.

26.

-7
and Virginia.

South Carolina,
that

serve

a diverse

group

They operate
of customers.

Restricting
Reserve

District

differences

eliminates

determine

the

possibility

the other

influences

one business

cycle

through

July

The use of monthly

1975).
of random,

from day to day.
reserves

being

aggregate
reserve

the analysis

(It

periods

positions

that

discount

may

transient
partly

far

average

factors

longer

individual

January
variables

1971

minimizes

that

lagged

affect

bank reserves

the

ameliorates

such as quarterly

aggregate

are examined over

(recovery-boom-recession:

based on two-week

over

any

of the

factors

on borrowing

roughly

impact

of confusing

activity.

The monthly

the

one Federal

administration

rationing--with

borrowing

", ,':-:

the sample to

in District-by-District

window--nonprice

numerous offices

effect

of required

deposits.)

It

than

those

relevant

or

yearly

intervals.

banks

into

does not

possibly

to

bank
Nor does

nonhomogenous

groups.

Variable Relationshi~s
The 28 variables
independent
and liability

of each other.

that

may explain
The behavioral

management postulates

that

borrowings

may not be

model of a bank's
strong

relationships

asset
exist

7The smallest
of these banking organizations
held over onequarter
billion
dollars
in deposits
as of December 31,1974.
West
Virginia's
small unit banks behave differently
from those sampled [23].

27.

among portfolio

ratios

themselves,

as well

as with

the monetary

climate.
A test
the

independent

analysis.8

analysis

9

numbers of variables
while
to

"distmce"

of
the

separates

each other--"
analysis

between

tionships

between

portfolio

items,

unrelated

thus

variables.
interest

It

patterns

rates,

interest
In this

the

explanatory

analysis

(not

the

strongest

differentials,

way,

knowledge
are is

are

dimensions.

rate

variables

traits

that

portrays

shows the mutually

large

related

in multiple

map" that

analysis

relate

highly
onto

angles"

a "road

and seasonality.

independent

factor
that

clusters
traits

in

of the data

Essentially,

It

that
present

dimensionality

coefficients

at right

is

technique

common influence

compact space.
correlation

among

made by factor

a multivariate

to each other.

it

Factor

is

reduces

It

multidimensional

orthogonal

the common relationships

patterns

of a relatively

together;

of

variables--multicollinearity--is

data set.

that

mutually

extent

the higher-order

any large

creates

the

Factor

reveals

to

for

relalocation,

of how

used to reduce

multicollinearity.

Methodolo8Z
The factor

8
Multicollinear
coefficients
(a singular
equations,
particularly

shown) suggests

operational

variables
generate indeterminate
cross-product
deviation
matrix)
those computed with an intercept.

specifications

estimates
of
in regression

9William Jackson,
Commercial Banking Performance and Structure:
A Factor Analysis
Appro~ch,
Federal Reserve Bank of Richmond, ~~orking
Paper 74-5,1974.

28.

to

test

rates

the initial

Bank Market

and interstate

models,

which

indicates

demand shifts

a smaller

dimensions
regression

analysis.

set

of possible

hypothesis
all

that

to-borrow"
only

theory

of the 1930's,

feel

random short-term

reserve

gives

the

borrowing:
differing

it
units

each variable's
influence

the

standardized

Y = bl

Y
-~
Sy

This

regression

the

"need"

"need-

borrow

caused by

"b"

+

B
1

(--+-)
Xl
sl

of each predictor

2

(--+-)
X2
s2

regression

bn ~

...+

of

the larger

is

its

These

model:

+ e

(8) coefficients
8

of

value

predictors.lO

of the

b2 X2 + ...+

in

their

absolute

coefficient,

coefficients

Xl

are presented

method of presentation

the

to the other

standardized
.,

the null

banks

of interpreting

The larger

relative

the

that

fo'r

in which

resembles

postulated

problem

of measurement.

into

a random process,

of the importance

the serious

lOThe computed

are converted

independent

against

of the regressions

coefficients.

on borrowing

tested

of unpredictable

results

comparison
avoids

are

the

It

pressures.

of standardized
direct

portfolio

used as input

hypothesis

which

the pressure

Technically,
the form

null

interest

space and over time.

They are

essentially
This

disregards

encompasses

These regressions

are zero.

when they

across

of variables

is

which

and the extended

"causality".

borrowing

coefficients

Model,

demand variations,

utilize

that

Line

of:~
8
n

(-+ )
Xn
sn

e'

where I's" represents
the standard
deviation
of a variable.
For each
increase
of one standard deviation
of an explanatory
variable,
the
standard deviation
of Y should increase
by that variable's
S. The
standardized
coefficients
of absolute dollar,
binary,
interest
rate,
and ratio
variables
can be compared directly.
Conventional
regression
("b") coefficients
cannot be compared directly
because of their
different
units of measurement.
Colin Clark et. al.,
Business and E~onomic
Forecasting:
An Econometric
Approa""Ch (HOmewood, Irwin:
1961),pp.
74-75.
Robert Ferber and P. J. Verdoorn,
Research Methods in Economics ~n~
Business
(New York:
Econometric
Theory

Macmillan,
(New York:

1962), p. 99.
Arthur
Wiley,
1964), p. 54.

S. Goldberger,

'7.
29.
coefficients

have been used in many economic

Bank Portfolios,

Size,

The test
Table

1 (p.

associated
loans.

cash,

Federal

minimizes
while

it

volatile

that

other
its

typified

Its

bank that
Regulation

and

funds soldll,
and bank-

organization
Federal

Government

tends

funds.

It

securities,

involves

relatively

profit-oriented

It

enterprise

returns.

equation

This

than

positively

Government deposits.

presumably

suggests

but rather

better

banking

structure

by fluctuating

grow larger,

banking

Federal

additions:

liability

an increasing

large

with

and unpledged

to borrow

proportion.

load

reserves

out loans.

seeks high

as they

are

in

Government deposits,

large

reserve

as a risk-accepting,

tend

model appears

Government securities,

borrowing

short-term

primary

seeks

This
not

associated

12 unpledged

a typical

deposits,

behaves

[41].

size.

That is,
to purchase

Line

borrowings

funds purchased.

cash items,

organization

Bank Market

1 shows that

They are negatively

vault
ing

Table

with

including

and Borrow!~g-

of the initial

30).

studies,

finding

organizations
smaller

ones.

that

large

proportion
that

may reflect
to match
The largest

banking

organizations

of their

they borrow

supply

banking

ability

and demand for
organizations

llAn institutional
consideration
is relevant:
sold Federal funds was regarded as violating
A during this period.

l2The factor
analysis
reveals
on different
factors,
reflecting

reserves

a decreasing

the longer-term
the

required

do

of very
funds
were

a borrowing
the spirit

of

that vault
cash and total
cash items
possibly
differing
bank characteristics.

30.

TABLE 1.
Borrowed Reserves/Required
Reserves
Explained
by Portfolio
Items Only*

Explanatory
Variable

Standardized
Regression
Coefficient

FFPUR

0.2672

t
Statistic
7.06

Significance
of t
0.0001

FFSOLD

-0.2399

-8.08

0.0001

VCASH

-0.2145

-4.81

0.0001

0.1703

3.51

0.0008

-0.1497

-5.65

0.0001

LOANS

0.1349

10.67

0.0001

USSEC

-0.1031

-3.33

0.0013

ASSETS

-0.0968

-2.65

0.0082

GOVDEP
CASH

R2

0.3242

F(8,1211)

72.63

*The intercept
has been suppressed to lessen multicollinearity.
The
t statistic
is computed under a null hypothesis
of a zero relationship.
Its significance
is the probability
that the relationship
is zero (unity
minus the confidence level).
Stepwise mode estimation
is used with the
significance
level for variable
acceptance or deletion
set at 0.05.
This regression
as a whole is significant
at the 0.0001 level.

Co

~-

31.

generally

able

to obtain

nondeposit

commercial

paper,

common stock,

terms

than

smaller

ones during

these

liability

and equity

organizations
borrowings
of

would

large

banks,

(Such a finding

The test
between states,
location

of the
appears

variables

each state,

the

These coefficients
Columbia

banking

South Carolina
the high

demand for

the positive

economic

activity

suggests

that

in

these

borrow

areas.l3

example,

l~or

1974 per

DC
MD
VA
NC
SC

Bank ~larket

""'~

associated

vary
of the
with

has been suppressed.
and District

of

somewhat more than similar
This

associated

finding
with

The Virginia

companies

capita

Lines

The coefficients

tend

may reflect
generally

high

coefficient
to borrow

slightly

incomes were:

$7,479
5,881
5,265
4,612
4,258

The Virginia
coefficient
appears to be more related
companies than to economic activity.

'",

typical

of multicollinearity.)

Maryland,

banks.

services

holding

factors

intercept

Virginia,

multibank

size-

intercept

Carolina

banking

banking

2 (p. 32).

conventional

and North

these

and Borrowin8.

in Table

organizations

that

Any positive

second hypothesis:

show that

extent

low cash holdings.

example

Size,

identify

since

funds,

to depend on other
relatively

notes,

on more favorable
To the

provide

be a clear

Portfolios,

period.

to borrow.

appears

would

Bank Location,

this

items

such as their

such as capital

and Eurodollars

not desire

relationship

funds

to bank holding

32.
32.

TABLE2.
Borrowed Reserves/Required Reserves
Explained by Portfolio
Items and Location*

Explanatory
Variable

Standardized
Regression
Coefficient

t
Statistic

Significance
of t
~

VA

0.6542

6.37

0.0001

MD

0.6263

7.02

0.0001

DC

0.6228

9.29

0.0001

SC

0.4484

8.27

0.0001

NC

0.4242

7.02

0.0001

FFPUR

0.2896

8.79

0.0001

FFSOLD

-0.2432

-8.43

0.0001

CASH

-0.2329

-7.95

0.0001

VCASH

-0.1503

-3.34

0.0012

SAVDEP

-0.1388

-2.98

0.0033

USSEC

-0.1232

-4.03

0.0002

GOVDEP

0.1201

2.44

0.0140

LOANS

0.0729

3.75

0.0004

R2

0.3815

F(13,1206)

57.22

*See the notes to Table 1.

'o~-.c..,

"..~".,,'

."

33.

more than similar
apparently
banks.

large banks, since multibank

seek out higher

risk

14 These coefficients

concentration

ratio

effects

asset size,

confirms
tionship,

confirms

This equation,

which is related

relationship

the expected positive

and portfolio

allocation

to the

to expectations.
composition

however, does not include
It does include

and borrowing.

are a highly

rela-

dependable low risk

The expected interaction

lowers the influence

a negative

This finding

deposit volatility-borrowings

since savings deposits
a bank.

either

the portfolio

to location.

between savings deposits

source of funds for

than non-affiliated

change, contrary

generally

shown in Table 1.

company banks

do not appear to be related

or to its

This regression

and returns

holding

between demand

of the loan/asset

ratio

taken by itself.

IAterest

Rates, Portfolio
The third

Lines vary with

Items, Seasonality.

regression

tests

the financial

of the seasonal borrowing
to enter

the equation

location

variables,

of interest-related

if

the hypothesis

climate.

privilege,

and Borrowi!!~~

In view of the importance
it

allows

they are significant.

whose effects
variables.

that Bank Market

quarterly
(It

variables

does not include

should be swamped by the effects
See Chart 3, p. 15.)

-

l4William Jackson, ~¥!.tiba~k ~ol~~ng ~ompan~~s and Bank Behav!~E,
Federal Reserve Bank of Richmond, Working Paper 75-1,1975.

,!:\

'""""""""'cL"""""",,,J'.;;",

I

I

34.
Table
climate

3 (p.

35) indicates

and bank decisions

ratios.

Its

explanatory

that

lessens

interaction

between

the importance

power is

fairly

high

the

financial

of some portfolio

for

a micro-banking

regression.
The Federal
now the

strongest

purchased
of

interest

ratio

to the factor

confirming

the

to appear

at the

competitors.
with

Finally,

a negative

collateral

against
No seasonal

considered

by aggressive

periods

banks

management (Federal
the demand for
f ace no explicit
economic

rather

that
funds

than

coefficient
banking

a strong

their

regression,
reappears,

slightly

securities
ability

with

organizations

than their

Government

appear

in

Table 3,

the equation.
of

lack
sold,

borrowings.
"seasonal

The level

is

this

and

smaller

again

to

~

serve

appear
as

reserves.

to enter

during

larger

despite

variables

eligible

on reserves

unpledged

borrowed

rate,

not be used in

window less

coefficient,

sold

borrowings.

bill

The size

the

discount

with

funds

is

FFLD, and USSEC are so correlated

LOANS could

for

differential

Federal

associated

analysis.)

tendency

rate

by the Treasury

(It,

that

interest

of borrowings.

represented

on borrowings.

loan/asset

discount

to be strongly

rates,

according

~

less

determinant

continue

influence
the

funds

tight

Rather,

money,

secondary
unpledged

need to borrow"
terms.

loan

particularly

reserves

These banking

temporal

although

U.S.

suitable
securities)

organizations
that

cannot

they were
demand pressures
those
for

felt
asset

describe
apparently
be described

in

34.

35.

TABLE 3.
Borrowed Reserves/Required Reserves Explained
by Portfolio
Items, Interest Rates, and Seasonality*
Standardized
Regression
Coefficient

Explanatory
Variable

t
Statistic

Significance
of t

9.20

0.0001

-

FFLD

0.3186

FFSOLD

-0.2735

-10.30

0.0001

FFPUR

0.1874

5.48

0.0001

TBILL

0.1747

6.91

0.0001

ASSETS

-0.0771

-2.46

0.0135

USSEC

-0.0581

-2.26

0.0225

R2

0.4499

F(6,12l3)

165.33

*See the notes to Table 1.
Time variables
are eliminated
selection
algorithm
since they are not significant.

,...~

~"."

~c."".._c,'

-"

by the

.--36.

Summary and Conclusi~?s
This
asset

develops

and liability

extended
it

study

to

of individual

management behavior.

analyze

provides

a model

bank reactions

the framework

bank risk-return

While

to regulation

to examine

the use

of the

of the portfolio

model are tested

by large

District

organizations

risk

Fifth

banking

model could

and other

Three versions

forces,

discount

using

be

window.

borrowing

as an indicator

of

acceptance.
Borrowing

oriented

banks

approach

volatile

liabilities,

deposits,

items,

asset
vault

lagging
borrow

cash,

longer-term

of this

rate.

behavior

of money,

the financial
further

raises

rate

market

monetary

aspects

the

as complementary,

funds

the

rises,

discount

supply

rate.

short

sold,

cash

as gross

increases

than

the
typically

This
as well

run response

Reserve policy.

window,

see

seem to use borrowings

not

low-

particularly

of loans,

Federal

discount

faster

organizations

dampens the

of the

banks thus

rates

banking

exceeds

in a way that

their

bank behavior

from lending

stimulates

and Government

They minimize

"aggressive"

return

purchased

Government securities.

economy to restrictive

Aggressive
purchased

funds

funds

profit-

They accept

such as Federal

Individual

when the absolute

presumably

mana~ement.

loans.

and unpledged

money climate

when the Federal

supply

portfolio

management accounts

discount

procyclical

a risk-accepting,

such as Federal

The incidence
when a tight

follow

to their

and extend

earning

'._-""""

this

substitute,

as the
of
(For

[32].)
and Federal
sources

funds

of short-

,,",c

I

-

37.

term

reserve

funds

adjustments.

apparently

Conservative
large

This
require
with
for

banks

"need"

can practice

deposit

study

borrowing

banks

appear

that

be considered

It

to

holding

that

of borrowing.

the

accounts.

to

utility

raise

short

count window,

banking

face

They do

organizations

are negatively

associated

special

treatment
since

with

within

required

influences

induced

the

is

no seasonal

time variables.

and deposit

run,

in areas

than similar

on

"Seasonal
reserve

spirit

banks in

pressures

of Regulation

less

A

activity

prosperous

also

areas,

suggests

that

more than similar

general

are inconsistent

consistent

risk

acceptance.

with

the

with

the profit-risk

window funds

at an elastic

"need"

of a bank.

of discount
the extension
particularly

indeed

economic

It

borrow

of their

relationships

model

of high

to assume risk.

indicator

They are

The availability

in

(FFLD).

of the year,

company organizations

These systematic

appears

loan

banks

borrow

banks--one

managerial

that

times

appropriate

demand inducements

unaffiliated

theory

volatile

them "seasonal."

are more likely

multibank

costs

when macro and micro-level

represent

does suggest

reflecting

assets

simultaneously

may thus

larger

suggest

at different

borrowing"

calling

that

Total

Nor does it

are examined

without

on these

They do not

from volatile

suggest

surveillance.

in borrowing

could

in their

asset management.

withdrawals

does not

special

borrowing

dependence

to borrow.

borrowing.

shifts

relative

depends on the difference

potential

not

Their

of

credit

during

may be the decisive

by these

ti~ht

banking

money periods.

inducement

for

rate

thus

organizations
An open
banks

dis-

to remain

38.
38.
38.
members of the Federal
predictable
tions.

by observing

This

study

Bank Market Lines,
in

these

Reserve System.

lines,

finds

The use of this

bank portfolio
that

as well

are important

banking

ratios

and money market

organizations'

as interstate
determinants

window may be

location

and interest-related
of

bank borrowing.

condi-

along
shifts

39.

APPENDIX:

_Bank Portfolios

and Profitabilit_~

The Bank Market
exists

Line

between nominal

liabilities.

This

banking

data,

necessary

profitability

connection

since

Fifth

expressed

and the

reserve

a strong

riskiness

can be tested

District

rate

as a potential

regression

is

variables

are entered

less

1,644

banks in

than

against

connection

of assets
independent

accounts

do not

cent of all

the equation

and
micro-

contain

for

the

Table

It

after

loan

indicates
losses
source

This

ratios.

the text,

only until

(then)

is

highly

composition.
that

This
that

contribution

their

historically

to

reports

high

sample includes

accordingly

except

from regulatory

loans

of funds

that

(Strong

multicollinearitYt
from being

It
time

by

interest

more than 11 per-

and savings

included

lower

the balance

broad

categories

of

shown in Appendix

returns

shows that

are a still
through

to four

regressiont

earn higher

than

demand deposits

related

This

are deducted.

cationt

more variables

in

is

banks.

Bank profitability
and liability

the

capital

portfolio

are derived

and 1971.*

insured

asset

of eighteen

The data

44 states

on bank equity

the same way as those

into

0.001.

of 1969,1970,

of return

function

computed in

R2 is

expensive

that

information.
The accounting

period

Model indicates

than investmentst

equity

capital

depositst
cost
sheet

is

even
a more

and by impli-

source
identity,

of funds.
prevents

in the regression.)

*The data were provided
by the Federal Deposit Insurance
Corporation.
They are described
in William
Jacksont Commercial Bank Regulation,
~trusture.
and Performanc~
(doctoral
dissertationt
University
of North
Carolina,
1974).

"""',1.. ",.""'-"

,."

ill_."

-"""

40.

40.
Appendix
Statistical

Characteristics

Table

of Profitability

Standardized
Regression
Coefficient

Explanatory
Variable

I
Equation

t
Statistic

Significance
of t

--

Loans/Total
Deposits

0.9868

28.00

0.0001

Investments/
Assets

0.7454

25.59

0.0001

-0.4409

13.87

0.0001

-0.3022

8.31

0.0001

Equity Capital/
Assets
Time and Savings
Deposits/Total
Deposits
R2

0.9500

F (4,1640)

7787.29

On the

side

while

profitability,

asset

on the

sources

of funds

a strong

positive

risk

and accounting

decrease

of

liability

profitability.

association
rates

the balance

exists

sheet,

side,

greater

longer-term
These findings

between

bank asset

loans
less

increase
risky

suggest
or

that

liability

of return.**

**Ex ~
realized
returns
from highly
risky loans declined
later
in the 1970Ts with a higher level
of loan losses.
The state
of nature--a
recession--generated
the unfavorable
portion
of the distribution
of returns for many banks.

-""",,"

,

41.
41.
REFERENCES
1.
2.

Aigner,
Dennis J., and Iolilliam R. Bryan, "The Determinants
of
Member Bank Borrowing:
A Critique,"
Journal
of Finance (December,
1968).
, and
, "A Model of Short-Run
---~
Bank Behavior,"

3.

Quarterly

Journal

-Of~o~~-

(February,

1971).

Alperstein,
Leslie M., "An Evaluation
of Some Determinants
of
Member Bank Borrowing,"
in ~~~~f~!s~l
of ~he ~ed~ral R~~~
Discount Mechanism, vol.
2.
Washington:
Board of Governors of
the Federal Reserve System, 1971.
---~-~

4.

Andersen, Leonall
C., and Jules M. Levine,
Market Conditions
as a Means of Short-Run
~ational
Banking Revie!. (March, 1967).

5.

, and Albert
E. Burger,
"Asset Management
.-and Commercial Bank Portfolio
Behavior:
Theory and Pract i ce, "
Journal
of Finance (May, 1969).
---~

6.

Berman, Peter I.,
"A Theoretical
and Empirical
Examination
of
Large Bank Use of the Discount
Facility
in the United States."
Doctoral
dissertation,
John Hopkins University,
1967.

7.

Betz, George W., "Member Bank Borrowing
from the Reserve Bank:
A Micro-Analytic
Analysis."
Doctoral
dissertation,
University
of Wisconsin,
1967.

8.

Bryan, William R., and Willard
T. Carleton,
of an Individual
Bank," Econometrica
(April,

9.

Cacy, J. A., "Determinants
of Member Bank Borrowing,"
Reserve Bank of Kansas City,
Monthly Review (February,

10.
Federal

,rA Test of Money
Monetary Management,"

"Short-Run
1967).

, "Reserve Adjustments--Some
Empirical
Reserve Bank of Kansas City,
Monthly Revi~!.

Adjustments

Federal
1971).
Findings,"
(June, 1971).

11.

Cagan, Philip,
"Interest
Rates and Bank Reserves --A
Reinterpre'tation
of the Statis tical
Association,"
in ~ssays on In ter~~~
~,
vol.
1, eds. Jack M. Guttentag
and Philip
Cagan.
New York:
National
Bureau of Economic Research, 1969.

12.

Carson, Deane, and Scott,
Ira 0., "Commercial Rank Attributes
and
Aversion to Risk,"
~~gking and Monetary Stud~e_~, ed. Deane Carson.
Homewood: Irwin,
1963.

,,",""""'-

I

42.

42.
13.

Crews, Joseph M., and Glenn C. Picou,
"Alternative
Reserve Concepts
as Operating
Targets in Monetary Policy Implementation:
Specification of the Structural
Model,"
Federal Reserve Bank of Richmond,
Working PaE.~~ 74-4,1974.

14.

De Leeuw, Frank,
Duesenberry
the
United

15.

~.
Stat~~.

DePamphi1is,

"A Model of Financial
2;l..,

~~_~~~ok!~~~
Chicago:

Donald

Behavior,"

g~r~~r1y
McNally,

Rand

M., ~~~§.~o~~~~~~c

-~;~nometric
1965.

~cgnometric

~~~~~~~r~~~~~_Cg~rc~a~
Bank A~_i~~tment Proce~~.
Federal Reserve Bank of Boston,
1974.
16.

17.

Dewald,
Respect
1970).

in James S.
Mode!~

Ana1ysis__~
Boston:

William
G., and G. Richard Dreese, "Bank Behavior \olith
to Deposit Variability,"
Journal
of Finance (September,
-

Emery, John T., "Risk, Return,
Banking ," Journal
of Financial

and the Morphology of Commercial
and Quantative
AnalY~i_~ (March,

1971).
18.

Francis,
Jack C.,
Englewood Cliffs:

and Archer,
Stephen H.,
Prentice-Hall,
1971.

19.

Frost,
Peter A., and Thomas J. Sargent,
"Money-Market Rates, the
Discount Rate, and Borrowing
from the Federal Reserve,"
Journal
of Money, Credit,
and Banking (February,
1970).

20.

Go1dfe1d,
Activ!t~.

Stephen M.,
Amsterdam:

21.
Member Bank Borrowing:
(June, 1966).
22.

Cg~.rc~a~_B~k
North-Holland,

Portfolio

~;~~vior
1966.

Ana1ysi~.

and Econom!c

, and Edward J. Kane, "The Determinants
of
An Econometric Model, " Journal
of Finance

, and
Member Bank Borrowing:
A Reply,'

,

Journal

, "The Determinants
of
of Finance (December,

1969).
23.

Harrison,
William B., and Paul A. Meyer,
Uncertainty,
and Institutional
Variables,"
Journal- (July,
1975).

24.

Hess, Alan C., "The Riskless
of Risk,"
Quarterly
Journal

25.

Hester,
Donald D., "Reflections
of Money, Credit,
and Banki£g

'M'","

..,

~,

...'c_.~,

"Banks'
Buffer
Accounts,
Southern Economic

Rate of Interest
and the Market
of Economics (August,
1975).
on the Discount
(May, 1970).

-.

Window,"

Price

Journ~l

.
43.
26.

, "An Empirical
Analysis
of Risk-Taking
by Firms
in the Savings and Loan Industry,"
in Karl Borch and Jan Mossin,
eds.,
Risk and Uncertainty.
New York:
St. Martins,
1969.

27.
Portfolio

Behavior.

28.
Bank Portfolios

(November,
29.

Hinderliter,
Position

Journal

, and James L. Pierce,
Bank Management and
New Haven:
Yale University
Press,
1975.

, and John
and Earnings
,"

F. Zoellner,
"The Relation
Between
Review of Economics
and Statistics

1966).
Roger H.,
"Market
Adjustments
of Large

of Finance

(March,

Access,
Uncertainty,
Commercial
Banks in

and Reservethe 1960's,"

1974).

30.

Hodgman, Donald R., "Member-Bank
of Finance (March, 1961).

31.

Jensen,
Journal

32.

Jones, David M., "A Review of Recent Academic Literature
on the
Discount Mechanism,"
in Reappraisal
of the Federal Reser~e ~i~co~t
Mechanism, vol.
2. Washington:
Board of Governors of the Federal
Reserve System, 1971.

33.

Kane, Edward J., and Burton G. Ma1kie1,
Deposit Variability,
and the Availability
Journal
of Economics (February,
1965).

34.

Kaufman, George G., "Deposit
0; Financial
and Quantitative

35.

Klein,
Michael
A., "The
Financial
Intermediation,"

Michael
(Autumn,

C., "Capital
1972).

36.
Theory,"

American

Borrowing:

Markets:

Variability
Analysis

Economics
Journal

Theory

A Comment,"

and

Journal

Evidence,"

Bell
-

"Bank Portfolio
Allocation,
Doctrine,"
Quarterly
and Bank Size,"
(December, 1972).

of Security
of Finance

Divisibility
(September,

, "Imperfect
Asset Elasticity
Economic Review (June, 1970).

Journal

and
1973).

and Portfolio

37.

Lapkin,
David T., and Ralph W. Pfouts,
"The Administration
of the
Discount Function,"
National
Banking Review (December, 1965).

38.

Luckett,
Dudley G., "Credit
Standards and Tight
~~ Money, Credit.
and Banking (November, 1970).

39.

Monho110n, Jimmie R., and James Parthemos,
"Administration
Discount Function:
A Comment," National
Banking Review
1966).

Money,"

~ourna1

of the
(September,

--z.--z.--:-

44.
40.

Moore, Basil J., A!;l1~~~roduction to the Theory of Fi?a~~.
York:
Free Press, 1968.

41.

Morgan, George E., ~_~~~k ~~~~~O~~o~Model of D~sc~unt ~~in~~
~orrowin~~.
Washington:
Reserve System, 1972.

New

Board of Governors of the Federal

42.

Partheoos, James, "The Discount t-lindow," in Instruments
.--~~
~oney Market, eds. J1mmie R. Monhollon and Glenn Picou,
Federal Reserve Bank of Richmond, 1974.

43.

Phaup, Marvin, "Regression Estimates of Bank Earnings by Portfolio
Category," Federal Reserve Bank of Cleveland, unpublished, 1974.

44.

Pierce,
Bulletin

45.

::

Ja~s 1.,
(August,

Polakoff,
Murray E., "Member-Bank Borrowing:
Journal of Finance~ (March, 1961).
~
Member Bank Borrowing~

48.
Member-Bank Borrowing:
(March, 1967).

A Rejoinder,"

, "Reluctance Elasticity,
Least Cost,
Journal of Finance (June, 1960).

and

~

, and William L. Silber,
"Reluctance and
Additional
Evidence,!' Journal of Finance

Pringle,
John J., "The Imperfect-Markets
Model of Commercial
Bank Financial Management," J_ournal of Financial and Quantitat!v_~
Analysi~

50.

Federal Reserve

:.

47.

49.

Bank Liquidity,"

, "The Monetary Mechanism: SomePartial
American Economic Review (May, 1964).

---

Relationships,"
46.

"Com~rcial
1966).

of the
Richmond:

(January,

1974).

Pyle, David H., "Descriptive
Theories of Financial
Institutions
Under Uncertainty, " ;!_ournal of Financial and Quantitative
Analy~~~
(December, 1972).

51.
52.

Russell, William R., "Commer
cial Bank Portfolio
Adj us tment s , "
American Economic Review (May, 1964).of the Discount Function-Shull, Bernard, "The Administration
~_.
Another View,"

53.

National

Banking Revi~w (March, 1967).

, "Report on Research Undertaken in Connection
;ith
a System Study," ~aEprais~l
0; t!!e Federal Reserve Discou~~
Mechanism, Vol.l
Washington:
Board of Governors of the Federal
R~serve System, 1971.

_k..."...,""".

I

-'..1.

54.

Silber,
William 1., :ortfolio
New York:
Holt,
1970.

Behavior

of Financial

55.

Tobin, James, Illiquidity
Preference
as Behavior
Review of Economic Studies
(February,
1958).

56.

Whittlesey,
Quarterly

C. R., "Credit
Policy
at the Discount
Journal
of Econorni£s (May, 1959).

Institutions.

Towards Risk,"

Window,"


Federal Reserve Bank of St. Louis, One Federal Reserve Bank Plaza, St. Louis, MO 63102