Full text of Economic Review (Federal Reserve Bank of Cleveland) : April 1965

View original document

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

APRIL 1965

THIS

ISSUE

Bank M a n a g e m e n t
o f Cash A s s e ts ...........

M a jo r Social
Insurance Trust
Funds— A S urvey . . .

FEDERAL

RESERVE

BANK

CLEVELAND

A d d itio n a l copies o f

the

EC O N O M IC

REVIEW

m ay be o b ta in e d from the Research D epartm ent,
Federal Reserve Bank o f C leve la nd , C le veland ,
O h io 4 4 1 0 1 . Permission is g ra n te d to re produce
any m a te ria l in this publicatio n.

Federal Reserve Bank of Cleveland
Cleveland, Ohio
ERRATA
ECONOMIC REVIEW, April 1965

Page 8

T h e word c a s u a l in the se co n d line sh o uld be c a u s a I .

A s c o rre c t e d , the f i r s t three l in e s sh ould re a d :

“ It wo uld thus s e em l i k e l y that an o b s e r v a b l e ,
and pe rhaps c a u s a l , r e l a t i o n s h i p e x i s t s b e
tween v a l u e s of the managed c a s h - t o t a l a s s e t . . .**

APRIL 1 9 6 5

BANK MANAGEMENT
OF CASH ASSETS

It is generally accepted that cash assets1,
which account for a substantial proportion of
total bank assets, constitute one of the least
productive uses of bank funds. Unlike loans
and investments, cash assets yield no money
income. By holding sizable amounts of cash a
bank incurs a cost equal to income that could
have been earned had such cash been used
to acquire earning assets.
B ecau se cash assets yield no return, it is
not surprising that bank managem ent over the
long run has tried to reduce the importance
of cash relative to total bank assets. This
article discusses some of the factors associ
ated with the attempt by banks to reduce the
proportion of total assets committed to cash,
with particular emphasis on the m anaged
1 In this article, "cash assets" are defined as the sum of
vault cash (sometimes referred to as currency and coin),
reserves maintained with regional Federal Reserve
banks, balances with other commercial banks in the
U. S., and cash items in process of collection.

cash 2 component over which bankers can
exert direct influence. The data used in this
article are for 1954 through the first half of
1963.3

COM POSITION OF CASH ASSETS
The volume and composition of cash assets
held by member banks of the Federal Reserve
2 For reasons discussed later in the article, the definition
of "m anaged cash assets" for the subperiod 1954
through mid-1960 differs from that for the subperiod
from mid-1960 through mid-1963. For the first sub
period, managed cash assets are defined as vault cash
plus correspondent balances with commercial banks in
the U. S. plus the difference between balances main
tained at the regional Federal Reserve banks and the
volume of required reserves (that is, excess reserves).
For the second subperiod, managed cash assets include
correspondent balances and excess reserves, which are
redefined as the difference between the total of balances
maintained at Reserve banks plus vault cash and the
volume of required reserves.
3 Analysis of the period since mid-1963 is not included
because comparable quarterly data are not available.

ECONOM IC REVIEW

System are influenced by both legal and
practical considerations. Regarding legal con
siderations, member banks are required by
law to maintain specified cash reserves against
their deposit liabilities. While reserve require
ments were originally intended to protect
depositors, this function is no longer the over
riding one. Rather, as it becam e obvious that
required reserves gave little additional liquid
ity to bankers, required reserves (or reserve
requirements) becam e primarily an instru
ment used by the central bank in the im ple
mentation of monetary policy.
Until Decem ber 1959, the total of legally
required reserves had to be kept on deposit at
the regional Federal Reserve banks. B ecause
of the compulsory nature of these reserves
and the possibility of adverse clearing b al
an ces at the Federal Reserve, deposits at the
Reserve banks were commonly maintained in
excess of amounts legally necessary, particu
larly by sm aller banks. Since 1959, member
banks have been able to count vault cash as
part of their legal required reserves, obviating
the need to maintain the whole of such b al
an ces with the Federal Reserve banks.4
A side from statutory requirements, and as
a practical consideration, operating needs
also make it n ecessary for bank managem ent

deposit withdrawals, and satisfying unfore
seen loan dem ands. B ecause vault cash may
now be counted as part of required reserves,
this also affects the total volume of cash kept
as vault cash.
Nearly all rural and suburban banks main
tain dem and b alan ces with city correspon
dents; city banks, in turn, also maintain cor
respondent relationships with similar institu
tions located throughout the country. Essenti
ally, the practice of correspondent banking
rests upon the uniquely Am erican system of
unit banking. Many relatively small country
banks look to city correspondents to provide
a host of services —ranging from investment
advice to help in designing a new banking
office—that they themselves could not eco
nomically furnish. Among other things, smaller
banks hope to participate with larger corre
spondents in the making of sizable loans for
which their own resources are inadequate.
Also, even though all member banks are free
to use the Federal Reserve's check clearing
facilities, some banks may, for one reason or
another, prefer to do some or all of their clear
ing through correspondents.5 This applies not
only to smaller banks, but also to large institu
tions. By providing these and other services,
correspondent banks incur costs that are

to hold some portion of total assets in various
forms of cash. A s with most individuals and

passed on to the recipient. To provide com
pensation, the user often maintains a cash

businesses, a part of these b alan ces must be
kept on hand as currency and coin (vault

b a la n c e —in the form of a dem and deposit —

cash). Such funds are necessary for the daily

5 For example, check clearing is sometimes faster by
using the facilities of a correspondent. Also, before
checks can be cleared through the Federal Reserve
System a certain amount of preliminary work must be
performed by the sending bank. Often the city corre
spondent will perform this for a country bank not having
either sufficient personnel or facilities.

conduct of banking business —cashing deposi
tors' checks, making change, meeting savings
4 The ability to count vault cash as part of required
reserves came in several stages, beginning in December
1959 and continuing into November 1960.

with each of his correspondents; the latter can

APRIL 1 9 6 5

then earn an interest income through profit
ably investing deposited funds.
Finally, becau se of the time required for a
bank to collect funds on checks drawn on
other institutions, some of its assets are always
tied up in cash items that are in the process of
collection. Though large in volume, these
items are not a very important source of
liquidity; a banker must assume, in the ab
sence of knowledge to the contrary, that as
checks are collected, they will be replaced by
a roughly similar amount of other such items,
so that the bank will always have some funds
tied up in the process of collection.

NATURE OF CASH ASSETS
C ash assets yield no income to their owner.
In fact, with the exception of correspondent
balan ces com pensating for services received,
an opportunity cost is incurred through the
holding of idle cash. That is to say, by not
acquiring interest-earning debt obligations or
not making appropriate business and other
loans, a bank foregoes earnings. Thus, bank
ers generally attempt to keep cash assets to a
minimum.
Bank m anagem ent can exert no control
over some items, such as required reserves
and cash items in process of collection. But

dent balan ces —the banker is able to exercise
discretion regarding both amount and com
position.6 The latter group of items comprise
m anaged cash assets. Such assets will be held
at minimum levels if maximization of profits,
consistent with prudence, is the banker's
prime objective.

BEHAVIOR OF CASH IN RELATION
TO TOTAL BANK ASSETS:
1 9 5 4 -1 9 6 3

For the years under consideration (1954
through the middle of 1963), the ratio of cash
assets to total assets of all member banks in
the United States declined by about 22 per
cent, or at an av erage annual rate of 2.46
percent. This secular decline is evident in
panel A of Chart 1. As the data for various
periods show, however, there was not a con
sistent pattern in the rate of decline. For

over other types of cash a sse ts—excess re
serves (as defined in each of the subperiods),
vault cash prior to mid-1960, and correspon

example, the average annual rate of decline
in the cash asset-total ratio after the first half
of 1960 substantially exceeded that in the
1954 through mid-1960 subperiod. Thus, the
average annual rate of decline expanded in
the later period to nearly three times the rate
of decrease in the earlier period.
As panel B shows, the component ratio of
m anaged cash to total assets likewise con
tracted over the entire 1954-63 time span.
The decline, however, was much larger (by

0 Some qualification is necessary at this point. To a
limited extent, management can affect the amounts of
cash items in process of collection and held as required
reserves. In the case of the latter, for example, it may be
possible to effect a change by altering the composition
of deposit liabilities. But opportunities for the exercise
of such discretion are distinctly limited. On the other
hand, practical considerations preclude bankers from

exercising unqualified discretion over the amounts and
relative distribution of managed cash assets. To illus
trate, some vault cash obviously must be kept on hand
and at least a part of correspondent balances is a neces
sary payment for services received. For analytical pur
poses such a classification is both meaningful and useful
if the limitations are in mind.

EC O N O M IC R E V IE W
1.
B E H A V IO R

O F SELECTED

B A N K ASSET R A T IO S

ALL M E M B E R

Po m I

BANKS

- U.S.

CA SH ASSET
T O TA L ASSET

P ercent

s /

R T S AE
AIO C L
111 11 1 III III
1954

1956

i 1 \ 1! I ' ! i i ! ! II 11 1
1958

196 0

1962

A V E R A G E A N N U A L RATE O F C H A N G E
195 4- 196 3

195 4- 196 0

1 9 6 0 -1 96 3

(first h a ll)

(first h a lf)

- 2 .4 5 8 %

-1.611%

- 4 .5 9 8 %

P«olB

M A N A G E D CA SH ASSETS

1954

1956

1958

196 0

1962

A V E R A G E A N N U A L RATE O F C H A N G E
1 954 -1 96 3

195 4- 196 0

(first h a lf)

- 5 .5 6 5 %

- 3 .0 8 3 %

-12.9 19 %

1 96 0 -1 96 3

M A N A G E D CASH A SSETS

234

about
times) than that registered by the
cash asset-total asset ratio. Also, beginning
with the second half of I9 6 0 , there w as a
pronounced increase in the av erage annual
rate at which the m anaged cash-total asset
ratio declined. From the first half of 1960
through the first half of 1963, the ratio fell, on
average, by nearly 13 percent a year against
only 3 percent in the earlier subperiod
(1954-60).
With the ratio of cash assets to total assets
contracting more slowly than the component
ratio, the proportion of m anaged cash to total
cash assets also declined (see panel C of
Chart 1). The in crease in the rate of decline
of m anaged cash relative to total cash b egin
ning in 1960 is, of course, the result of differ
ing rates of chan ge between the total of cash
assets and the m anaged cash element.
It is important to recognize that the pro
nounced declines in the ratio of m anaged
cash to total assets carry primary responsibil
ity for pulling down the proportion of total
cash to total assets. As Table I demonstrates,
the ratio of nonm anaged cash (cash assets not
subject to the bank's own control —required
reserves and cash items in process of collec
tion) to total assets declined at a much lower
av erage annual rate than the ratio of m anaged
cash to total assets, and even considerably

1954

1956

1958

1960

1962

A V E R A G E A N N U A L RATE O F C H A N G E
1 9 5 4 -1 96 3

1 9 5 4 -1 96 0

(first h a lf)

- 4 .0 0 6 %

- 1.63 8 %

- 9 .6 9 6 %

Source o f d a ta :

196 0- 196 3

B o a r d o f G o v e r n o r s o f t h e F e d e r a l R e s e r v e Sy s t e m

less than the decline in the cash asset-total
asset ratio. Moreover, becau se of the relative
ly low proportion of weight of m anaged cash
in the total of cash assets,7 the long-term de
cline in the ratio of cash assets to total assets
was due largely to the sizable reduction in the
7 From panel C of Chart 1 it can be seen that the man
aged cash component accounted for approximately 20
percent of total cash assets.

APRIL 1 9 6 5

TA BLE I
Changes in Selected Bank A sset Ratios
(A v e ra g e Annual Rates)
Managed Cash Assets
All Member Banks— U.S.

1954-1963
1954-1960
1960-1963

Cash Assets

Nonmanaged Cash Assets

Total Assets

—
—
—

5 . 57 %
3.08
12.92

—
—
—

2.4 6 %
1.61
4.60

1. 39 %
1.11
— 2.10

—

Note: In all cases, measurement is from first half of the year beginning the period to the first half of the closing year.
Source: Board of Governors of the Federal Reserve System

component m anaged cash-total asset ratio.
The in creased pace at which the cash
asset-total asset ratio declined in the 1960
through mid-1963 subperiod also should be
attributed largely to the behavior of the man
ag ed cash-total asset ratio. While there was
some increase in the average annual rate at
which the ratio of nonm anaged cash to total
assets declined, it was not com parable to the
increase in the rate of decline of the m anaged
cash-total asset ratio.8
It appears evident from the foregoing that
bankers have been successful in reducing the
proportion of total bank assets committed to
discretionary cash holdings. Such su ccess is
even more marked when it is recognized that
a portion (whose size is statistically unknown)
of m anaged cash assets lies outside of the
banker's span of control (see footnote 6). The
m agnitude of the long-term decline and in
crease in the rate of decline of the m anaged
cash-total asset ratio stands out in sharp con
trast to the more moderate rates of d ecrease
registered by the nonm anaged cash-total
8 In any case, the increase probably just compensated
for the sharp decline in the weight assigned to managed
cash in the total of cash assets (see panel C of Chart 1).
As the dynamic factor, any loss of weight not compen
sated for elsewhere would have retarded the rate at
which the cash asset-total asset ratio declined.

asset ratio.
Thus, the following conclusions em erge:
(1) the ratio of cash assets to total assets de
clined over the 1954-63 period; (2) after the
first half of 1960, the rate of decline increased;
(3) long-term reductions in the proportion of
cash to total assets are primarily due to sizable
declines in the component m an aged cashtotal asset ratio; and (4) the in creased rate of
decline in the ratio of cash assets to total
assets is explained basically by the more pro
nounced decline in the component m anaged
cash ratio.
From these conclusions two questions
em erge. First, what factors accounted for the
long-term decline in the m anaged cash-total
asset ratio? Second, what factors explain the
in creased p ace of decline since the first half
of 1960?

FACTORS DETERMINING THE
BEH AVIO R OF THE RATIO OF
MANAGED CASH TO TOTAL
ASSETS
As opportunity costs of holding discretion
ary cash rise and actual dollar outlays in
crease relative to revenues, bank m an age
ment will seek to further minimize cash b al
an ces and to employ existing funds profitably.
7

ECONOM IC REVIEW

It would thus seem likely that an observable,
and perhaps casual, relationship exists b e
tween values of the m anaged cash-total asset
ratio and the following factors (variables):
levels of short-term interest rates, the propor
tion of time to dem and deposits, and the ratio
of total expenses to revenues. A reasonable
hypothesis is that declines in the ratio of man
aged cash to total assets are associated with
increases in (1) the level of short-term interest
rates, (2) the proportion of time to dem and
deposits, and (3) the ratio of total expenses to
total revenues. Before turning to data to
verify or reject this hypothesis, a discussion of
the associations that are involved would be
helpful inasm uch as what are being looked
for are relationships expressing causality —
not simply observable associations.
Increases in short-term interest rates make
the cost of keeping idle cash b alan ces more
expensive; that is to say the holder of such
b alan ces forfeits the opportunity of gaining
additional revenue by not acquiring interestearning assets. Thus, when short-term rates
increase, bankers can be expected to further
reduce m anaged cash balances. At the same
time, when expenses increase relative to
revenues, thereby reducing profits, bankers
are additionally motivated to seek out new
ways of cutting costs and raising earnings,
with the investment of formerly idle cash
representing one alternative.
More attention must be devoted to the re
lationship between the m anaged cash-total
asset ratio and the proportion of time to d e
mand deposits. Banks, of course, pay interest
on time deposits while they do not pay interest
on dem and deposits. And with all things
equal, this would make time deposits more
8

costly and less profitable than dem and de
posits. The evidence available, however, is
not conclusive as to the relative cost and
profitability of the two types of deposits.9 This
is not totally unexpected for things are not
always equal; a number of other considera
tions related to time deposits, such as lower
reserve requirements and smaller handling
costs, must be taken into account and these
may offset increases in costs due to interest
payments. R egardless of which point of view
is closer to being correct, there are at least
two reasons for expecting a rising proportion
of time to dem and deposits to result in declines
in the m anaged cash-total asset ratio. First, on
the basis of casu al thought and in the absen ce
of detailed cost studies, bankers generally
believe time deposits to be more costly than
dem and deposits, and consequently will strive
to have more of their assets in earning form.
Second, time deposits have traditionally been
considered as behaving in a less volatile and
more predictable manner in contrast to the
behavior of dem and deposits; this view has
justified maintaining lesser amounts of im
mediate liquidity (cash).
A gainst this background, consideration can
be given to the association between the fac
tors discussed above (short-term interest rates,
the ratio of time to dem and deposits, and the
proportion of total expenses to total revenues)
and the behavior of the m anaged cash-total
asset ratio.
9 See Crosse, H. D., M a n a g e m e n t P o lic ie s fo r Com
m e rcia l B an ks, (Englewood Cliffs, N. J.: Prentice-Hall,
Inc., 1962), Chapter V and "The Cost of Demand
Deposits," N ew E n g la n d B u sin e ss R eview , Federal
Reserve Bank of Boston, October 1961, for two diver
gent views concerning relative profitability. Interestingly,
neither source takes a particularly rigid position in the
matter.

APRIL 1 9 6 5

SELECTED B A N K IN G

AND

ALL M E M B E R

1954

1956

E C O N O M IC V A R IA B L E S
BANKS

1958

- U .S.

1960

1962

P ercent

tES
TOT A L E)(P EN D IT U F
TO TAL

III

1954

' •

1956

II!

1958

RFV

ENUE

1 1!

1960

III

1962

P ercent

TOT AL Tl V\ D EPO S ITS
E
TOT A 1 n F M A M n nppr»QiT<
......

111

III

...

— % '■
"’V

1954

f
K ni

III

1956

III

::
1958

III

1960

III

1962

P ercent

3 -MONT H BIL L RA 'E (N«w I s s u e s )

III
1954

! i
1956

1958

III
196 0

III
1962

S o u r c e o f d a t a : B o a r d o f G o v e r n o r s o f t h e F e d e r a l R e s e r v e S ys te m

A s shown in Chart 2, the ratios of both time
to dem and deposits and total expen ses to total
revenues10 rose steadily but unevenly over the
1954-63 period. O n the other hand, short
term interest rates fluctuated in a generally
cyclical manner, and the ratio of m anaged
cash to total assets declined secularly.
It is to be expected that relatively low values
of the ratio of m anaged cash to total assets, as
a general matter, will be associated with
relatively high rates of interest and high pro
portions of time to dem and deposits and ex
penses to revenues. Conversely, relatively
high values of the m anaged cash to total assets
ratio may be associated with relatively low
interest rates and low proportions of time to
dem and deposits and expenses to revenues.
These expectations are to some extent sup
ported by the data plotted in Chart 2. Stronger
support, however, is found in Chart 3 where,
in a scatter diagram , the various factors under
consideration are plotted against the m anaged
cash asset-total asset ratio. In each case, an
inverse relationship is indicated (downward
sloping to the right); relatively high values of
the m anaged cash-total asset ratio are gener
ally associated with relatively low values of
the explanatory variables. As Chart 3 shows,
these negative relationships are particularly
evident with respect to the ratio of time to
dem and deposits and the proportion of ex
penses to revenues.
10 For purposes of this study, the value of the total
expense-total revenue ratio attributed to a given year is
actually that of the previous year. To illustrate, for 1960
the value of the ratio in 1959 is used. The reason for this
is that in the course of a year bank management may
only be aware of the prior year's expense-revenue
relationship. It is this relationship that influences cash
management policies in the current year.

R E L A T I O NS H I P of SELECTED V A R I A B L E S to M A N A G E D C A S H ASSETS - T O T A L ASSET R A TI O
ALL M EM BER B A N K S - U.S.

Ratio

of Managed

Cash

Assets

to Total As se ts

3- M o n t h Bill Ra te on N e w Issue s

http://fraser.stlouisfed.org/
Sour e of d ta
Bo
Federal Reservec Bankaof: St. a r d o f G o v e r n o r s
Louis

R a t io of Tim e D ep o sits to D e m a n d D ep o sits

R a t io of Tota l E x p e n s e to Total R e v e n u e
o f t h e F e d e r a l R e s e r v e S y s te m

R E L AT I ON S H I P of SELECTED V A R I A B L E S to M A N A G E D C A S H ASSETS - T O T A L ASSET RATI O
ALL M EM BER B A N K S - U .S.

_J _____________ I______________I______________I_____________ I______________I______________I______________ 2%
1.0 %

1.5 %

2 .5 %

3 .0 %

3 .5 %

4 .0 %

4 .5 %

3 -M o n th B ill R a te on N e w Issu es
Ratio
of M anaged
Cash
Assets
t Total A ssets
o

60%
61%
http://fraser.stlouisfed.org/
S ource of d a ta : B o a rd of G o v e rn o rs
Federal Reserve Bank of St. Louis

2 .0 %

R a tio of Tim e D e p o sits to D e m a n d D ep o sits

62%

63%

64%

65%

66%

R a tio of T o ta l E x p e n s e to T o ta l R e v e n u e
o f t h e F e d e r a l R e s e r v e S ys te m

67%

68%

69%

70%

When the various scatter points in Chart 3
are connected and labeled in a time sequence
(this is done in Chart 4), it becom es evident
that a permanent downward shift occurred in
the m agnitudes of the m anaged cash-total
asset ratio between the second and third
quarters of 1960. (The time period is identi
fied by a circle on the chart.) The second
quarter of 1960 thus marks a break in the
data; in succeedin g periods given values of
the explanatory factor are associated with
substantially lower values of the m anaged
cash-total asset ratio. The nature of the break
in the data is perhaps most clear in the panel
showing the relationship between the 3-month
bill rate and the m anaged cash-total asset
ratio.
B ecau se of the downward shift, the degree
of decline in each of the subperiods (1954
through the first half of 1960 and the second
half of 1960 through mid-1963) is actually
less than indicated in Chart 3. As an illustra
tion, Chart 3 reveals a rather pronounced
negative relationship between values of the
m an aged cash-total asset ratio and those of
the ratio of time to dem and deposits (declines
in the former are associated with in creases in
the latter). However, turning to Chart 4, and
centering attention on the two subperiods, the
negative association is not nearly so pro
nounced. This observation also holds, more or
less, for the other two sets of relationships.
C onsidering the two subperiods individu
ally, it is evident that declines in the ratio of
m anaged cash to total assets are associated
with in creases in short-term interest rates, the
proportions of time to dem and deposits, and
total expen ses to total revenues. Furthermore,
it seem s reasonable to conclude that long
term declines in the m anaged cash-total asset
ratio are in part explained by in creases in the
12

above-mentioned factors.11
Having accounted for the declines in the
ratio of m anaged cash to total assets in each
of the subperiods, the next step is to explain
the sharp increase in the av erage annual rate
of decline since mid-1960. The observed
increase can be explained in part by the sam e
factors discu ssed earlier (that is, short-term
interest rates, the ratio of time to dem and
deposits, and the proportion of expen ses to
revenues).12
The data in Table II present supporting
evidence. The statistics m easure the percent
decline in the ratio of m anaged cash to total
assets associated with a one-percent increase
in the values of each of the explanatory fac
tors.13 Thus, for exam ple, for the 1954 to mid11 On the basis of statistical analysis, approximately 68
percent and 57 percent of the variation in the behavior
of the managed cash-total asset ratio, for the 1954 to
mid-1960 and mid-1960 to mid-1963 subperiods,
respectively, is explained by the three factors cited
herein.
12 It should also be pointed out that a number of impor
tant institutional changes have taken place in the 1960's,
which, according to some observers, have "revolution
ized" banking. Such developments as greater use of
certificates of deposit, a more active Federal funds
market, and, as a more general matter, the tendency of
many bankers to look upon their liabilities as a source
of liquidity have been associated to some extent with the
increased rate of decline exhibited by the ratio of man
aged cash to total assets since mid-1960. However, it is
not certain whether these developments have been
causal influences or merely institutional responses to
the factors discussed in the text.
13 The statistics in Table II have been estimated from
the following equations:
(a) (M anaged Cash-Total Asset Ratio) =
ai + Bi (Short-Term Interest Rates)
(b) (Managed Cash-Total Asset Ratio) =
a 2 + B2 (Demand-Time Deposit Ratio)
(c) (M anaged Cash-Total Asset Ratio) =
a3 + B 3 (Expense-Revenue Ratio)
For each subperiod, estimates have been made on the
basis of data pertinent to the period in question.

TABLE II
Percent Decline in the Ratio of
M anaged Cash A ssets to Total A ssets
A sso ciated With a 1 % Increase in :
First Quarter

Third Quarter

1954 through

1960 through

Second Quarter Second Quarter
1960

1963

(a) Short-Term Interest Rates

0 .1 2 9 %

1 .0 55%

(b) Demand-Time Deposit Ratio

0.546

0.864

2.393

2.845

Source: Estimated by the Federal Reserve Bank of Cleveland

1960 subperiod, a one-percent in crease in
short-term interest rates (e.g., from 3 .0 0 per
cent to 3 .0 3 percent) resulted, on average, in
a 0 .1 2 9 percent decline in the m anaged cashtotal asset ratio. The data thus indicate the
responsiveness of b an k ers—as m easured by
the percentage decline in the ratio of m anaged
cash to total assets —to chan ges (increases) in
the factors that are important in the maximiza
tion of bank profits. For the period since mid1960, Table II clearly reveals a marked im
provement in m anagem ent's responsiveness
to advan ces in short-term interest rates and to
increasing proportions of time to dem and
deposits and expenses to revenues. W hereas
for the earlier subperiod, a one-percent in
crease in short-term interest rates was associ
ated with a decline of only 0 .1 2 9 percent in
the value of the m anaged cash ratio, a similar
increase in the later subperiod was associated
with a 1.055 percent decline.
Improved responsiveness, however, only
partially explains the marked increase in the
av erage annual rate of decline exhibited by
the ratio of m anaged cash to total assets since
mid-1960. Complementing and possibly in
fluencing responsiveness was the sharp break,
or downward shift, in values of the m anaged
cash-total asset ratio that took place between
the second and third quarters of 1960. In this

connection, it should be noted that this rather
sudden and sustained shift resulted from per
mission granted member banks by the Board
of G overnors of the Federal Reserve System
in D ecem ber 1959 to count vault cash as part
of their legally required reserves. Until then,
legal reserve requirem ents could be satisfied
only by maintaining deposit b alan ces at re
gional Federal Reserve banks. At the same
time, operating needs m ade it n ecessary for
m anagem ent to keep on hand sizable amounts
of vault cash.
Prior to permission to count vault cash,
excess reserves were the difference between
the volume of reserves maintained with the
regional Federal Reserve banks and the vol
ume of required reserves. In succeedin g
periods, however, it was no longer necessary
that all of required reserves be kept on de
posit with the Reserve banks; vault cash,
either in part or in total, could t e substituted
for Federal Reserve balances. Excess reserves,
therefore, becam e the difference between the
sum of cash b alan ces maintained with the
Federal Reserve plus vault cash and the
amount of legally required reserves. To the
extent that bankers chose to keep required
reserves in the form of vault cash, instead of
as deposits at the regional Reserve banks, the
former ceased being a m anaged cash asset
(that is, an asset whose behavior is subject to
m anagem ent's discretion). Thus, a modifica
tion in the form in which reserve requirements
were satisfied cau sed a reduction in the
volume of m anaged cash assets.
This is seen in Table III. In the 1954-60
subperiod, excess reserves av eraged about
$ 577 million, while the vault cash component
of m anaged cash assets averaged about $2,151
million. In the su cceedin g subperiod, excess
reserves rose to $61 9 million. However, as
13

TABLE III
M anaged Cash A ssets
(A v e ra g e Q u a rte rly D o lla r Volume)

(1)

(4)

(2)
(3)
(in Millions of Dollars)
Balances
With Other

Managed Cash

Vault

Commercial

Assets

Cash

Banks in U.S.

Reserves

Excess

9,586

2,151

6,858

577

8,001

—

7,382

619

524

First Quarter 1954
through Second Quarter 1960
Third Quarter 1960
through Second Quarter 1963
Net Change*

— 1,585

— 2,151

*Not additive because of rounding.
Source: Board of Governors of the Federal Reserve System

m anagem ent effected sizable reductions in
b alan ces at the Reserve banks, all of vault
cash cam e to be counted towards required
reserves in addition to serving its traditional
function as an operating asset. O nce bankers
choose to satisfy reserve requirements in this
fashion, vault cash ceased being an asset
subject to m anagerial control. The decline in
m anaged cash assets resulting from the vault
cash provision was partially offset by increases
of $ 5 2 4 million in balan ces with other com
m ercial banks and $41 million in excess
reserves. The former probably reflected the
additional use of correspondent services as
the volume of check clearings rose with over

vision on in creased m anagerial responsive
n ess to increases in short-term interest rates
and the proportions of time to dem and d e
posits and expenses to revenues. While evi
dence pointing to in creased responsiveness
was presented earlier, no attempt was m ade
to explain why. The primary reason for not
attempting to account for m anagem ent's in
creased responsiveness is that the cau ses are
not sufficiently apparent or easily validated.
However, by being able to count vault cash as
part of required reserves it does seem that
bankers were provided with additional flexi
bility, which in turn enabled them to be more
responsive to factors tending to constrict

all econom ic activity. In any event, as cash
which no longer n eeded to be maintained at

profits. Before the provision, values of the
m anaged cash-total asset ratio had already

the regional Reserve banks found more profit
able investment, m anaged cash assets de
clined by slightly more than $ 1 ,585 million.

been reduced to relatively low levels; add i
tional declines would have been difficult to
bring about. The vault cash provision ap p ar

CONCLUDING COMMENTS

ently altered the nature of the environment
within which cash m anagem ent had previ

It would be extremely difficult to document
precisely the influence of the vault cash pro
14

ously operated and provided m anagem ent
with a new set of opportunities to further
curtail unprofitable m anaged cash assets.

APRIL 1 9 6 5

MAJOR SOCIAL INSURANCE
TRUST FUNDS
“ T ru st fu n d s are c re a te d w hen C o n
g re ss d e sig n a te s c e rta in re c e ip ts to be se t
a sid e —in t r u s t —fo r sp ecified p a y m e n ts
on p r o g r a m s . . . T ru st fu n d s m a y be u sed
on ly to fin an ce the a c tiv itie s fo r w hich
they are d e s ig n a te d . . . M o st o f the m a jo r
tr u s t fu n d s fin an ce in su ra n c e -ty p e a c tiv
itie s su c h a s so c ia l se c u rity (o ld -a g e ,
su rviv ors, a n d d isa b ility in s u r a n c e ), u n

e m p lo y m e n t in su ra n c e , F e d e r a l e m p lo y
ees' re tir e m e n t, a n d v e te ra n s’ life in s u r
an ce . . . R e c e ip ts o f the tr u s t fu n d s com e
fro m so u rc e s r e la te d to th eir a c tiv ity . . .
In m a n y c a se s, tr u s t fu n d r e c e ip ts exceed
e x p e n d itu re s. T he excess is a c c u m u la te d
in reserv es to m e e t fu t u r e p a y m e n ts a n d
is u su a lly in v e ste d in U. S . se c u r itie s to
earn in te r e st fo r the f u n d .” 1

A SURVEY
The foregoing is a highly simplified sum
mary of a vast area of Federal Government
finance that is often overlooked in discussions
of fiscal policy, debt management, financial
flows in credit and capital markets, private
spending and saving decisions, and the like.
There are more than 60 U. S. Government
trust funds and investment accounts, with
assets totaling in excess of $60 billion as of
June 30, 1964. In asset size alone, the com

cial banks.
A further indication of some of the financial
m agnitudes —or the inflows and outflows of
funds —involved in trust fund operations is
su ggested by the fact that expenditures of all
trust funds combined accounted for nearly
one-fourth of the Federal Governm ent's cash
payments to the public during fiscal year
1964 ($28.9 billion com pared with $ 1 20.3
billion). The corresponding ratio of trust fund

bined funds are larger than the three largest
life insurance com panies or the eight largest
manufacturing corporations in the U. S. Re
flecting their asset size as well as their invest

receipts to total Federal receipts was slightly
higher ($30.3 billion com pared with $115.5
billion). These relationships give some idea of
the significance of trust fund operations with
in the fiscal policy complex of the Federal

ment policies, the trust funds hold nearly onefifth of the total Federal debt, or almost three
times as much as that held by all nonfinancial
corporations and about the sam e as commer

1 The B u d g e t in B rief, Fiscal Year 1966, Executive Office of
the President, Bureau of the Budget, Washington, D. C.,
p. 52.

ECONOM IC REVIEW

Government. Furthermore, receipts of some
$ 3 0 billion in fiscal year 1964 meant that
trust fund inflows approxim ated the sales
revenues of the eighteen largest m erchandis
ing firms in the U. S.
A gainst this background, it is the purpose
of this article to survey the operations of
selected Federal trust funds and to discuss
some of the implications of their program s.
Attention is confined to the seven largest
social insurance trust funds, which account
for approxim ately 85 percent of total assets of
all trust funds combined. In addition to form
ing the major part of the total trust fund com
plex, these seven funds also share a common
socio-economic tie of carrying out the func
tions of national social insurance programs.

BACKGROUND
Background information on the seven trust
funds is presented in summary form in Table I
in the A ppendix. In addition to the financing
provisions of the respective program s, infor
mation is included on the statutory authority
under which each operates. It should be
noted that the financing provisions of the var
ious social insurance program s are changed
frequently. (See Tables II through V in the
Appendix.) Further chan ges are expected to
be m ade during the current session of the
C ongress. For exam ple, the President has re
quested that chan ges be made this year in the
O ld-A ge and Survivors' Insurance program,
involving increases in benefit payments, tax
contributions and the taxable earnings base
(effective in 1966), and the incorporation of
a m edical care for the aged program.
16

The O ld-A ge a n d S u rv iv o rs’ In su ra n c e
program is probably the most important of the
various Government social insurance pro
gram s, particularly b ecau se it affects more
people than any other program . The O A SI
program usually is considered together with
the D isa b ility In su r a n c e program, since
they are financed with combined employment
taxes and cover the sam e workers. In fact, the
two program s together (OASDI) are often re
ferred to as the social security program . At
the end of 1964, more than 90 percent of all
U. S. workers in paid employment (including
self-employment) were covered by the social
security program .2 Those excluded were
Federal employees, certain em ployees of state
and local governments, and m edical doctors
in private practice. Many of those excluded
from social security are covered by other
Federal or state and local government retire
ment and disability plans.
For exam ple, the C ivil Serv ice R e tir e
m e n t a n d D isa b ility program provides re
tirement and disability benefits for em ployees
of the Federal Government. The Vice Presi
dent, M embers of C ongress, and con gres
sional em ployees may come under the pro
gram at their own discretion and are referred
to as Members. The R a ilro a d R e tir e m e n t
A cco u n t, set up before railroad workers
were affiliated with the social security pro
gram, makes monthly benefit payments to
retired and disabled railroad em ployees and
their dependents and also provides lump-sum
death benefits.
2 Including railroad employees. Although more than 90
percent of workers were covered, a slightly smaller
proportion was actually in the program. Coverage for
some workers is voluntary.

APRIL 1 9 6 5

The u n e m p lo y m e n t in su ra n c e program
was established by the states under the aegis
of the Federal Government. The program is
also financed by an employment tax, but one
that applies to em ployers only. Tax receipts
are collected by state agencies, and then the
reserves of the various state program s are
transferred to the national Unemployment
Trust Fund. Unlike the social security pro
gram, there is variation in coverage and
benefits within the unemployment program,
reflecting the diversified nature of that
program .3
The other two trust funds discussed here
involve still another kind of insurance. The
veterans' life insurance program s provide life
insurance with a maximum value of $10 ,0 0 0
to qualified veterans. Both the U n ited S t a t e s
G o v e rn m en t L ife In su ra n c e and the N a
tio n a l Serv ice L ife In su ra n c e programs,
which issued policies to veterans of World
W ar I and World W ar II, respectively, were
closed to new entries after 1951.4 The volume
of benefits, dividends, and refunds by these
trust funds has been declining as fewer poli
cies are in force.

SOURCES OF FUNDS
The receipts and expenditures of the trust
funds both reflect and influence the behavior
of the economy. That is to say, since revenues
of the trust funds represent a drain upon
current income and expenditures represent a
3 The various states have different rules and criteria
regarding coverage, duration and size of benefit pay
ments, and financing.
4 Congress recently passed legislation reopening the
NSLI program for one year beginning May 1, 1965 to
World War II and Korean War veterans.

supplement to current income, the interaction
of both sides of the accounts can have major
economic effects. Particularly becau se the
sam e people who make payments to social
insurance program s are not necessarily those
who receive payments, a number of economic
cross-effects are possible. These are discussed
in a later section.
Revenues of the seven trust funds are de
rived from a variety of sources: employment
taxes, policy premiums, interest and profits
on investments, "financial in terch ange", and
m iscellaneous sources. Employment taxes
and investment earnings are clearly the major
ones, as can be seen from Table I.
E m p lo y m e n t T axes. Five of the seven
major social insurance program s are primar
ily financed through employment taxes paid
by employees, employers, or both.5 Tax re
ceipts are taken here to include deposits re
ceived by the states and transferred to the
Federal trust funds. These five funds derived
$ 2 2 .3 billion, or 89 percent, of their total
income during fiscal year 1964 from such
taxes. With the exception of the Railroad
Retirement Account, each of the tax-support
ed funds currently derives 80 percent or
more of its revenues from employment taxes.
It might be noted that the Railroad Retirement
Account was also primarily tax-supported
prior to the development of the financial inter
change program, discussed on page 18.
Growing dem and for liberalization and
expansion of benefits has m ade it n ecessary
to increase social insurance tax contributions
5 Technically, employee and employing agency pay
ments to the Civil Service Retirement and Disability Fund
are not taxes; they are contributions. However, they are
considered as taxes for the purpose of this article.

ECONOM IC REVIEW

from time to time. Thus, for exam ple, in order
to establish adequate reserves for current
operations and reserves that were also suffi
cient to provide for substantially increased
claim s during periods of economic decline,
unemployment tax rates were raised in 1961
and 1962. The history of financing the five
tax-supported funds since their inception is
shown in A ppendix Tables II through V;
the tables also include financing projections
according to the provisions of current laws.
P olicy P r e m iu m s. In contrast to the retire
ment and unemployment plans, the two
veterans' life insurance funds derive impor
tant support from premiums paid by policy
holders. During fiscal year 1964, the two
veterans' life insurance funds com bined d e
rived $ 4 9 7 million, or 70 percent of their
revenues from policy premiums. However,
the U SG L I fund depended upon premiums
for only about one-third of its revenue, while
the NSLI fund, which insures younger veter
ans, received a larger proportion of prem i
ums. In fact, there has been a shift over time
from policy premiums to earnings on invest
ments as the major source of revenue for the
U SG LI fund.
I n te r e s t a n d P ro fits on In v e stm e n ts. As
a source of funds, earnings on investments
accounted for only $ 1 .6 billion, or 6.1 per

funds combined com prise a sm aller percen
tage of total revenues than they did in the
past.
F in a n c ia l In te rc h a n g e . In general, finan
cial interchange represents annual net trans
fers of funds between the OASDI program
and the Railroad Retirement Account. The
arrangem ent is intended to place the social
security trust funds in the financial position
they would have been in if railroad em ployees
were, and always had been, covered under
social security. W hen the financial inter
chan ge program was originated in the early
1950's, the net adjustment was in favor of the
social security funds. However, since 1957,
annual financial flows have been heavily in
favor of the Railroad Retirement Account, with
the reimbursements equivalent to the differ
ence between benefits that would have been
available on the b asis of com bined credits
under both retirement program s and benefits
actually paid on the basis of OASDI credits
alone. In fiscal 1964, financial interchange
totaled $ 422 million and m ade up approxi
mately 35 percent of the Railroad A ccount's
revenue.
O th er S o u rc e s. All other receipts of the
major social insurance trust funds accounted
for about $ 3 0 9 million, or slightly more than
1 percent of total revenues, during the 1964

cent, of total revenue for the seven trust funds

fiscal year. Other sources include advan ces

during fiscal year 1964. Such earnings are a
secondary source of income for the larger
trust funds, fluctuating narrowly accordin g to
the volume of security holdings. While rela

from the general revenues of the U. S. Treas
ury and interfund transactions.

tively small, earnings on investments do repre
sent a proportionately large source of revenue
for the life insurance funds. G enerally speak
ing, earnings on investments for the seven
18

USES OF FUNDS
G enerally, revenues of the trust funds are
used to make benefit payments, to pay adm in
istrative expenses and m iscellaneous expen
ditures, and, if funds are left over, to build up
the reserves of the funds.

TABLE I
Receipts and Expenditures of M ajor Social Insurance Trust Funds
Fiscal Y e a r 1 9 6 4
(m illions o f do lla rs)
Receipts

Old-Age and
Survivors’
Insurance

Disability
Insurance

$14,335
1,167
539
—
—
3
$16,043

$1,057
86
68
—
—
—
$1,211

593
—
130
422
—
47
$1,192

$14,579
—
300
403
3
$15,285
+ 759

$1,251
—
70
19
—
$1,341
— 130

$1,092
—
11
—
35
$1,138
+ 54

Employment T a x e s ...........................................
Deposits by S t a t e s ...........................................
Earnings on Investm ents.................................
Financial Interchange......................................
Policy Prem ium s................................................
O t h e r ...................................................................
T O T A L .........................................................

Railroad
Retirement
Account

United States
Government
Life Insurance

National
Service
Life Insurance

842
3,042
213
—
—
191
$4,288

$1,974
—
420
—
—
62
$2,456

—

$ 34
—

$176
—

16
—
$ 51

482
6
$664

$ 134
2,703
22
—
848
$3,707
+ 582

Civil Service
Retirement and
Disability

$1,318
—
—
—
—

$ 73
—
—
—
—

$588
—
—
—
—

$1,318
+ 1,138

$ 73
— 22

$588
+ 76

Unemployment
$

Total, Seven
Trust Funds
$18,802
4,295
1,580
422
498
309
$25,906

Expenditures
Benefit Payments................................................
Withdrawals by S t a t e s ..................................
Administrative Exp enses..................................
Financial Interchange......................................
O t h e r ...................................................................
T O T A L .........................................................
Additions to Reserves (— ) .............................

$19,036
2,703
404
422
885
$23,450
+ 2,455

Figures may not add to totals due to rounding.
Source: U. S. Treasury Department

TA B LE II
A ssets of M ajor Social Insurance Trust Funds
June 3 0 , 1 9 6 4
(millions o f dollars)
Trust Fund
Old-Age and Survivors’ Insu rance...................
Disability Insurance................................................
U n e m p lo y m e n t.....................................................
Civil Service Retirement and Disability! . . .
Railroad Retirement A c co u n t.............................
National Service Life In s u r a n c e * ...................
United States Government Life Insurance*
T O T A L .........................................................

Total Assets

Cash

Total Government
Securities

$19,726
2,264
6,858
14,386
3,859
6,303
1,075
$54,471

$1,421
126
40
107
92
13
1
$1,799

$18,305
2,138
6,818
14,279
3,766
5,783
956
$52,046

Figures may not add to totals due to rounding.
fPartially estimated.
*Other assets, including policy loans, liens, and receivables, totaled $508 million for NSLI and $1 18 million for USGLI.
Sources: U. S. Treasury Department, Veterans' Administration, U. S. Department of Health, Education, and Welfare

Public Issues of
Government
Securities
$3,506
361
1,888
788
798

—
—

$7,341

Special
Issues of
Government
Securities

Special Issues as
a Percentage of
Total Government
Securities

$14,799
1,777
4,931
13,491
2,968
5,783
956
$44,706

80 .8%
83.1
72.3
94.5
78.8
100.0
100.0
85.9%

ECONOM IC REVIEW

B en efit P a y m e n ts. The primary use of
funds is to make benefit payments to persons
covered by the respective program s. Such
payments are m ade to retired and disabled
workers (or their survivors), unemployed
workers, and policyholders (or their bene
ficiaries) under the veterans' life insurance
program s. The $ 21.7 billion of benefit pay
ments, including withdrawals by states, a c
counted for 93 percent of the total expendi
tures of the seven large trust funds during
fiscal year 1964. (See Table I.)
Benefit payments under OASDI accounted
for two-thirds of all expenditures of the seven
funds during the 1964 fiscal year. A pproxi
mately two-thirds of O A SI's $ 1 4 .6 billion in
benefit payments m ade during fiscal year
1964 accru ed to retired workers. The rem ain
ing payments were m ade monthly to depen
dents of retired or deceased workers or took
the form of lump-sum death benefits. The
disability insurance portion of the OASDI
program m akes benefit payments to disabled
workers un d er a g e 65 (and their dependents)
who are covered by the program . About 80
percent of this program 's $1.3 billion of bene
fit payments in fiscal 1964 was received by
working-age people who were permanently
and totally disabled.
A number of observations may be m ade
about the volume of benefit payments under

the differing ag e groups of veterans insured
under the two program s. Benefit payments of
the unemployment program were relatively
low by historical standards in 1964, reflecting
the high level of economic activity in the
nation.
A d m in istra tiv e E x p en ses. Four of the
seven large trust funds must pay their own
administrative costs; the other three are fi
nanced from general revenues of the T reas
ury. Administrative expen ses totaled $404
million in fiscal 1964, or 1.7 percent of total
expenditures. The percen tage relationship
for that year was generally similar to that
which has prevailed over time, reflecting the
relatively low administrative costs associated
with the various program s.
O th er E x p e n d itu re s. All other expendi
tures, including the financial interchange
program discussed earlier, accounted for 5
percent of total uses of funds in the 1964
fiscal year. Although financial interchange
accounted for a substantial proportion of the
Railroad Retirement A ccount's revenue, it was
not a correspondingly large percen tage of the
expenditures of the OASDI trust funds b e
cau se of the larger volume of spending under
the social security program .
A d d itio n to R eserv es. Receipts of the trust
funds not used in the current accounting

individual program s in fiscal 1964. In the
case of both the disability insurance portion
of the OASDI program and the U SG LI pro
gram s, benefit payments exceeded total re

period are added to reserves of the funds and
invested in public debt securities. Two of the
seven large funds (Disability Insurance and
USGLI) experienced declines in their reserves

ceipts in fiscal 1964. The NSLI program paid
out $ 5 8 8 million in benefits, dividends, and
refunds, as com pared with $73 million dis
tributed by U S G L I—a relationship reflecting

in fiscal year 1964 b ecau se expenditures ex
ceeded receipts. The other funds added
approxim ately $2.5 billion, or 9.5 percent of
total revenues, to reserves during the year.

APRIL 1 9 6 5
1
.
T R A N S F E R P A Y M E N T S a n d SELECTED C O M P O N E N T S
S e a s o n a ll y Ad ju st e d A n nu a l Ra tes
B i l l i o t s o f d o lla r s

45
Q u a rte rly

OLD AGE, S U R V I V O R S ’,
AN D DIS AB IL IT Y
IN S UR AN CE BENEFITS
STATE U N E M P L O Y M E N T
INS UR AN CE BENEFITS
VET E RA NS ’ BENEFITS

OTHER

'53
S ource o f d a ta :

’54

’ 55

’56

'57

U.S. D e p a r t m e n t o f C o m m e r c e

Accum ulation of reserves helps to fulfill
program or trust fund objectives. Reserves at
the least must be adequate to supplement
receipts in order to ensure current payments
and provide a basis for making expanded
payments in periods of economic decline
when benefits usually rise. While the ac c u
mulation of trust fund reserves would seem to
be contrary to the view that argu es for "d e
ficit spen din g" at the national level to help
stimulate the economy, proponents of con
tinually increasing reserves argu e that a
decline in the reserve base would harm the
actuarial soundness of the trust program s. It
is generally accepted that total reserves need
not be large enough to cover total liabilities,
but should be sufficient to meet current pay
ments, at least for a while. The seemingly
appropriate test for financial soundness is that
future income should support the volume of
anticipated disbursements.
Accumulation of trust fund reserves results
in a net withdrawal of funds from the current
income stream of the economy. In contrast,

deficits and drains on reserves provide addi
tions to the current income stream. The O A SI
trust fund has had deficits during five of the
last seven fiscal years, while the disability
trust fund has run a deficit in the last two
years. In fact, the Civil Service fund is the
only one of the seven funds that has not had
a deficit in recent years.

INVESTMENT POLICIES
The assets of the seven major trust funds
are summarized in Table II. A s indicated
earlier, the $ 5 4 .5 billion of assets of the seven
funds represented about 85 percent of total
assets of all U. S. Government trust funds as
of June 30, 1964. Of the seven, the O A SI
trust fund is the largest, with the Civil Service
Retirement and Disability fund ranking sec
ond. These two, combined, account for well
over half of the total assets of the seven trust
funds shown in the table.
In all cases, holdings of U. S. Government
securities constitute nearly all of the assets
held by these funds; cash balan ces and other
21

ECONOM IC REVIEW

assets represent only about four percent of
total assets. About 86 percent of the G overn
ment securities held by the trust funds are
special issues, that is, nonmarketable securi
ties sold exclusively to the trust funds. As
Table II shows, the seven funds held $ 44.7
billion in special issues as of June 30, 1964.
The Secretary of the Treasury, who is m an ag
ing trustee of each of the seven trust funds,
invests the funds' assets primarily in special
issues to protect the funds from price fluctua
tions of regular, marketable Treasury issues.
A ccording to law, each of the funds must
purchase special issues based on prescribed
interest rate formulas; the formulas have been
revised in the past few years in order to obtain
higher rates of interest.
Some trust funds also are authorized to
purchase other securities that are fully gu ar
anteed by the Government, and some funds
are even permitted broader investment p rac
tices. For exam ple, the U SG LI Trust Fund
may purchase nonguaranteed securities is
sued under authorization of the Federal Farm
Loan Act of 1916, and both veterans' life
insurance funds invest a small proportion of
their resources in policy loans.6

6 As of June 30, 1964, the USGLI fund held 9.7 percent
and the NSLI fund 8.8 percent of assets in policy loans.
The USGLI is the only major trust fund that has ever held
any securities other than public debt obligations. From
1925 through 1944, this fund held varying amounts of
Federal farm loan bonds. Such bonds made up a major
portion of the fund's investment portfolio during the
1930's. When in April 1964 the fund purchased a $25
million Federal Land Bank bond (a nonguaranteed
security) bearing interest at 4 percent, it represented
the first time since 1944 that the fund had held any
securities other than special issues of public debt
obligations.

SOME ECONOMIC EFFECTS
Benefit payments by the seven social insur
ance program s form a significant part of the
transfer payments component of personal
income. The seven program s accounted for
approximately 60 percent of the $ 3 8 .4 billion
of total transfer payments in calen dar year
1964, with the OASDI program alone a c
counting for more than 4 0 percent of the
total. Chart 1 shows the dollar volume of total
transfer payments and selected components
from 1953 through 1964. The chart indicates
continuous growth in total transfer payments
as well as a shifting pattern am ong the differ
ent types of payments. Although total transfer
payments were approxim ately two and a half
times as great in 1964 as in 1953, benefits
under the OASDI program were more than
four times as great, w hereas veterans' bene
fits were only about one-third larg er.7
Transfer payments com prise a relatively
small share of total personal income, but the
proportion has in creased in recent years. For
exam ple, in 1953 —a year that saw the begin
ning of a recession when payments would
tend to be la r g e r —transfer payments a c
counted for only 5 percent of personal in
come. In 1964, they amounted to 8 percent of
personal income. A considerable part of the
increase in transfer payments relative to per
sonal income reflects the rise in the number
of people under the social security program.
Although it is difficult to separate cyclical
movements from trend, a major importance of
transfer payments in the income stream is
based on their contracyclical movements
7 The disability insurance program was not in effect
until 1957.

APRIL 1 9 6 5

relative to personal income. Thus, while trans
fer payments have continued to increase
regard less of the stage of the business cycle,
the rate of increase has accelerated during
recession periods. During the 1953-1964
period, transfer payments increased by an
average of 4.1 percent during each recession
quarter but averaged only a 1.5 percent
quarterly increase during the expansion
phases of the business cycle (see Table III).
Within total transfer payments, the unemploy
ment insurance benefits component reveals
the most apparent contracyclical movements
(see Chart 1).

2.
T R A N S F E R P A Y M E N T S a n d W A G E S a n d S A L A R IE S as a
P ER CE NT o f P E R S O N A L I N C O M E
S e a s o n a ll y A d ju st e d A n n u a l Rates
P ercen t

69
68
67
66

’5 3

’ 55

S o u r c e o f data *.

TA BLE III
A ve ra g e Q u arterly Rate of Growth
in Personal Incom e and Selected Com ponents
1 9 5 3 -1 9 6 4
(percent)
Personal
Income

Transfer
Payments

Wages
& Salaries

Recession............................

+ 0 .7

+ 4.1

+ 0 .5

Recovery and Expansion

+ 1 .5

+ 1 .6

Source: U. S. Department of Commerce

The increased proportion of personal in
come accounted for by transfer payments and
the contracyclical activity of the social insur
ance program s does not completely depend
upon discretionary Government action; some
of it is ''autom atic” or "built-in". Thus, the ebb

’ 57

’59

U.S. D e p a r t m e n t o f

’ 61

’ 63

’ 65

C om m erce

than are other personal income components,
but in c o m b in a tio n , the two components
fluctuated by no more than 0 .5 percentage
points a quarter during the 1953-1964 peri
od. This reflects the fact that cyclical declines
in w ages and salaries have tended to be offset
by increases in transfer payments. In contrast,
during the recovery phases of the business
cycle, transfer payments have leveled off and
w ages and salaries in creased in response to
the improvement in economic activity.
The social security and other retirement
program s act as automatic stabilizers when
more persons choose early retirement during

and flow of payments under various pro
gram s reflect in part chan ges in economic
activity, particularly changes in the level of

recession periods, thus causing an increase
in transfer payments. Continually broader

income. This relationship is illustrated by the
behavior of transfer payments and w ages and

these retirement program s, however, obscure
the portion of total payments that is a response
to the state of economic conditions. In con
trast, the contracyclical effects of the social

salaries. Transfer payments and w ages and
salaries are plotted as a percent of total per
sonal income in Chart 2. Both components
are more responsive to cyclical movements

coverage and liberalization of benefits under

insurance program s are readily apparent in
the patterns of unemployment insurance
23

ECONOM IC REVIEW

benefits, as shown in Chart 3. Peaks in the
volume of these benefits usually have oc
curred either at the trough of a recession or
in the early months of expansion, with such
benefits increasing as more persons lose their
jobs and begin to draw unemployment com
pensation, and declining as these persons
return to their jobs when economic activity
improves.
Thus far, the economic effects of the social
insurance program s have been discussed
mainly in terms of trust fund expenditures,
primarily in the form of benefits paid. Trust
fund receipts also have economic significance,
as is indicated by Chart 4, which shows the
dollar total of personal contributions for all
social insurance program s during the 19531964 period. Rather than moving contracyclically, personal contributions have been
characterized more by a steady upward
trend. Some of the growth in total receipts
has been due to expanded coverage of work

ers under the social security program, but the
rising level may be traced even more directly
to legislative in creases in the amount of tax
contributions (indicated by the arrows on the
chart).
Increases in either OASDI tax rates or the
taxable earnings b ase have been passed by
C ongress primarily to help maintain actu
arial balan ce of the trust funds. Somewhat
coincidentally, such legislative increases
have often occurred during or only a few
months preceding the three most recent re
cessions. As a result, these legislative actions,
perhaps inadvertently, have tended to dam p
en the contracyclical effects of the social
insurance program s.
Finally, the social insurance program s have
an economic impact in that they bring about
some redistribution of income. The effect of
such redistribution is to shift potential pur
chasing power from employed workers (who
pay the taxes) to persons who are unem

S TA TE U N E M P L O Y M E N T IN S U R A N C E BEN EFIT S

PERSONAL C O N T R IB U T IO N S fo r SOCIAL IN S U R A N C E

S e a s o n a ll y A dju ste d A n nu a l R ate s

S e a s o n a ll y A d ju s t e d A n n u a l Rates
l l l l i o i s of d o lla rs

’53
'S3

’ 55

S ource of d a ta :

'57

'59

'61

U.S. D e p a r t m e n t o f C o m m e r c e

’ 63

’ 65

Source of d a ta :

’ 55

’ 57

'5 9

’ 61

’ 63

'65

U .S . D e p a r t m e n t o f C o m m e r c e

NOTE: Legislative increases in taxes indicated by arrows.

APRIL 1 9 6 5

ployed, retired, or disabled (who receive the
benefit payments). G enerally speaking, re
cipients of transfer payments are likely to
spend more of their total income than are the
employed people who are contributors to
social insurance program s. To the extent that
income is redistributed in this way, total con
sumer spending is increased and economic
activity is stimulated.
However, one additional factor to be con
sidered in evaluating the effects of income
redistribution is the fact that social insurance
taxes are regressive in nature. That is, by
setting a maximum income subject to the tax,
persons whose incomes rise above the tax

able earnings b ase pay a smaller percentage
of their total income in social insurance taxes
than do those whose incomes fall below the
tax base. To give an exam ple, an employee
with an annual gross income of $ 4 8 0 0 pays
3 5 8 percent of his salary, or $174, in OASDI
/
taxes. In contrast, a person with an annual
income of $ 4 8 ,0 0 0 also pays $174, or less
than 0 .4 percent of his income to the social
security program . Thus, any possible in
creases in purchasing power evolving from
the redistribution of income is partially blunt
ed by the regressivity of the tax burden and
as a result, may not have as great an eco
nomic impact as otherwise.

A P P EN D IX TA B LE I
Background Inform ation on M ajor So cial Insurance Trust Funds
Name of
Trust Fund

Statutory
Authority

Major Source
of Revenue

Insurance

Act Amendments of 1939

3 Vs

Self-employed'

5.025

$4,800 per year

Employees
Employers

Employment tax

Social Security

'A
'A
3
A

4,800 per year

Disability Insurance2

Act Amendments of 1956

Employment tax

Self-employed

Unemployment3

Social Security Act of 1935

Employment tax

Employers of 4 or more

Civil Service Retirement4

Civil Service

and Disability

Retirement Act of 1920

Employees
"Members"
Employment tax

Employing agencies

3.1

3,000 per year

6 Vi
7Vi
6Vi or 7V2

10,000 per year

Employees
Act of 1937

8 Vs

Employers

Railroad Retirement
Railroad Retirement Account5

Taxable
Earnings Base

3 3s %
/

Employers

Social Security

Rate

Employees
Old-Age and Survivors’ '.2

Paid by

8 Vs

Employment tax

Employees' Representatives

Policy premiums

Policyholders

National Service

Life Insurance Act of 1940

United States Government4

World W a r Veterans

Interest

Debtors

Life Insurance

Act of 1924

Policy premiums

Policyholders

450 per month

National Service

Life Insurance

16'A

n.a.

n.a.— not applicable.
1 The OASI trust fund superseded the Old-Age Reserve Account as of January 1, 1940. The Old-Age Reserve Account was established by the Social Security Act of 1935.
2 The OASI and Dl trust funds are jointly financed with the tax rate equivalent to 3Va percent for employees and employers and 5.4 percent for the self-employed.
3 A 2.7 percent tax credit is allowed for moneys paid into a state unemployment insurance program.
4 “Members” refers to the Vice President, Members of Congress, and congressional employees. The employing agencies match contributions by employees or "Members".

5 A Railroad Retirement Account was originally established in 1934 by an act that was declared unconstitutional in 1935. Litigation also prevented a 1935 Act from
becoming effective.
6 This fund was established in 1919 by an amendment to the W ar Risk Insurance Act of 1914. The 1924 statute effectually replaced the prior legislation. Recently
the major source of revenue for this fund has been interest on investments in U. S. Government securities, policy loans, liens, and premiums paid in arrears.
Sources: Social Security Act, Civil Service Retirement Act, Railroad Retirement Act, Veterans' Administration, U. S. Treasury Department

A P P EN D IX TA B LE II
Financin g P ro vision s of the O ld -A g e and
S u rv iv o rs’ and D isa b ility Insurance Trust Funds
Time
Period

Taxable
Earnings Base

Employees-Employers
Tax Rate (Percent)

A PPEN D IX TA B LE IV
Financing Provisions of the C iv il Service
Retirement and D isab ility Fund
Self-Employed
Tax Rate (Percent)

OASI

1951 - 1953

. . .

1954
1955- 1956
1957- 1958
1959
1960- 1961

OASI

—

1 7i

—

3,600

1/2
2
2
2
2 'A

—

2 'A
3
3
3
3Va
4 Vs

—

4,800

—

4,800

. . .

3,000

1937- 1949
1950

4.325

5.025

.
.

. . . .

1962

5.825

Va
Va

4,200
4,200

4,800

1963- 1965

. . .

1966- 1967

. . .

—
—

'A
\A
'A
'A
!'A
'A
'A

4,800

. .

1968 and after

Taxable Earnings Base

4,800

4Ve

Basic Salary

2Vi

27i

July 1926-June 1942

...................

Basic Salary

3'/2

July 1942-June 1948

...................

Basic Salary

August 1920-June 1926

. . . .

—

July 1948-October 1956

. . .

Basic Salary

—

November 1956-present*

. . .

Basic Salary

6'/2

7Zi

Va
Va

6.525

Source: U. S. Department of Health, Education, and W elfare

*Beginning July 1957 contributions by employees and members have been matched
by employing agencies.
Source: U. S. Civil Service Commission

A PPEN D IX TA B LE V
Financing Provisions of the R ailroad
Retirement A ccount

A P P EN D IX TA B LE III
Financin g P ro vision s of the Unem ploym ent
Trust Fund
Time Period
1935-1955
1956-1960

Paid By
. . .
. . .

1 9 6 1 ........................
1 9 6 2 * ...................
1 9 6 3 * ...................

Tax Rate (Percent)
Employees
Members

$3,000

3,600
. . .
. . .

Time Period

Employers of
8 or more

Monthly Taxable
Earnings Base

Time Period

Tax Rate (Percent)
Employees- Employees’
Employers Representatives

5Vi

$3,000 per employee

3.0

Employers of
4 or more

3,000 per employee

3.0

Employers of
4 or more

3,000 per employee

3.1

Employers of
4 or more

3,000 per employee

3.5

Employers of
4 or more

3,000 per employee

3.35

...........................................

$300

1940-1942

...........................................

300

1943-1945

...........................................

300

3'A

6'A

1946 .........................................................

Taxable Earnings Base Tax Rate (Percent)

1937-1939

300

3 '/2

1947-1948

300

11 '/2

...........................................

1 9 4 9 - 1 9 5 1 ...........................................

300

1952-June 1954

300

12'/2

.................................

July 1954-May 1959

........................

350

6 Vi

12'/2

June 1 9 5 9 - 1 9 6 1 .................................

400

13'/2

1962-October 1963

........................

400

7'A

14'/2

November 1963-1964 ........................

450

7'A

]4 '/i

1965

450

8 Vs

16 'A

...........................................

450

85
/s

*Tax rates for 1962 and 1963 were increased by the Temporary Extended
Unemployment Compensation Act of 1961.

17'A

1968 and a f t e r .................................

450

9'/8

1814

Source: U. S. Treasury Department

Source: U. S. Railroad Retirement Board

1964 and after.

Employers of
4 or more

3,000 per employee

3.1

1966-1967

Fourth Federal

Reserve

District

Full text of Economic Review (Federal Reserve Bank of Cleveland) : April 1965

FRASER