Full text of The Velocity of Money

View original document

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

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

The
Velocity of
Money
George Garvy and Martin R.Blyn

Federal Reserve Bank of New York

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Second Printing, Revised March 1970

COPIES OF THIS BOOK are available from the Publications Department, Federal Reserve Bank of New York, New York, N. Y. 10045,
at $1.50 per copy. Educational institutions may obtain quantities
for classroom use at 75 cents per copy.

Library of Congress Catalog Card Number 77-101695

Foreword
R. GARVY'S "Deposit Velocity and its Significance", which
was published by the Federal Reserve Bank of New York
in 1959, has been out of print for some time. What had started out
as an updating of the original booklet led to a completely revised
monograph which embodies some of the continuing research on
monetary economics going forward at the Federal Reserve Bank of
New York.
Since 1959, important developments have taken place in the
monetary process as a result of an interaction of rising opportunity
costs of money and of advances in cash-saving techniques. Even
more radical technological changes in payments mechanisms are
being widely discussed. Also, quantitative and other research on the
relationship between monetary and real processes has advanced
considerably in the decade since the publication of Mr. Garvy's
booklet. The present volume, the result of a collaboration by Mr.
Garvy and Professor Martin R. Blyn of California State College,
Dominguez Hills, reviews and evaluates recent developments with
respect to the velocity of money in the light of the new experience
and knowledge.

ALFRED HAYES
President
New York City
October 1969

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Acknowledgments
The authors wish to acknowledge the unfailing assistance of Miss
Abigail M. Cantwell, who edited the manuscript, prepared it for the
printer, and saw it through the press. Mr. Thomas Draper helped
with the initial editing, Mr. Patrick P. Kildoyle prepared the tables,
Miss Evelyn Katz checked the tables and all the references, Messrs.
John H. Hendrickson and Sigurds Vidzirkste prepared the charts,
and the Misses Kirsti Laandi and Margaret Riebel deserve particular
credit for patient typing of the successive drafts.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Contents:
Page
1. Introduction
2. The Demand for Money
Definition of Money
Money Substitutes
The Role of Financial Markets
Synchronization of Receipts and Expenditures
Factors Tending to Reduce the Demand for Transactions Money
Activation of Idle Balances
Compensating Balances
Models of Demand for Money

9
14
17
20
23
24
25
27
28
32

3. The Flow of Payments
The Flow of Consumer Payments
The Flow of Business Payments
Payments related to production
Intracorporate transfers of funds
The Flow of Savings and Investment Funds

4. The Measurement of Velocity
Income Velocity
Transactions Velocity

47
49

37
39
39
40
43

Sector Velocities

53
58

5. The Statistical Record
Cyclical Fluctuations
Long-run Changes

59
59
62

6. Factors Affecting Velocity
Compensating Arrangements
Economizing on Corporate Cash
Investing Temporarily Redundant Corporate Cash
Economizing on Personal Balances

67
67
68
74
76

7. Implications of Recent Changes in Velocity

Appendix I: Composition of the Money Supply

Appendix II: Share of Currency in the Money Supply

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

113

Page
CHARTS
Chart I: Money as a Percentage of Total Liquid Asset Holdings, 1953-68

Chart 2:Income and Transactions Velocity, 1919-68

Chart 3: Variants of Income Velocity, 1948-68

Chart 4:Bank Debits,Demand Deposits, and Rate of Turnover
in "Outside" Centers, 1919-68

Chart 5: Rates of Turnover of Demand Deposits, 1943-68

Chart 6:Income Velocity of Money in Three Post-World War II Business Cycles
Chart 7:Ratio of Transactions Velocity to Income Velocity, 1919-68

60
65

Chart8:Measures of the Relationship Between Money Payments
and Cash Holdings of Nonfinancial Corporations

TABLES
Table I: Cyclical Changes in Gross National Product,
Money Supply, and Income Velocity, 1946-69

Table 2: Cyclical Changes in Demand Deposits,
Debits, and Transactions in Velocity, 1945-69

Table 3: Selected Short-term Debt Outstanding

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

The Velocity of Money

1. Introduction
There are many ways of looking at the role of money in our economy. One important facet of monetary analysis is the relationship of income to the stock of money
—the velocity of money. The relationship between cash balances and the flow of
income or total payments may be defined and measured in a variety of ways, but
both theory and experience suggest that the relationship is not stable. The money
supply may change significantly, while the flow of income or the volume of payments associated with it remains almost unchanged; the reverse may also occur.
The resulting changes in the velocity of money are of considerable interest to
analysts as well as to policy makers.
Indeed, changes in the rate of money turnover may be as significant in determining policies intended to achieve appropriate levels of spending as the quantity of
money itself, however measured. The historical record shows,for instance, that the
occurrence of a 0.2 point increase from one year to another (e.g.,from 4.4 to 4.6)
in the rate of income velocity is by no means exceptional. Such an increase would
be equivalent to an expansion in the stock of money2 of approximately 4 percent
(or roughly $7 billion in 1968).
Since holders of cash can adjust their holdings—allowing for delays, frictions,
and inertia—to changes inflows, the demand for cash balances has been the subject

lAs occurred between 1966 and 1967.
2 For reasons given in Chapter 2, we prefer a definition of money limited to currency in the hands
of the public and demand deposits, except interbank and United States Government deposits.'

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

of many empirical studies of the monetary process. The question of the stability
of the relationship of income flows to the stock of money is of particular significance in assessing alternative objectives for, and efficiency in, the conduct of monetary policy. The relationship is very complex for income as well as transactions
velocity of money. To be significant, its analysis must go beyond interpretation of
statistical data.
Because money serves several distinct functions, changes in the money stock in
relation to the flow of goods and services or to the total volume of payments may
be traced to influences on the demand for money as a means of payment and
for other functions. One of the difficulties in analyzing the money stock is that the
demand for money balances is determined not only by economic motives, which
can be rationalized or at least categorized in terms of theory, but also by purely
technical factors, which affect the efficiency of making payments, locally and at
distant points. Over the years, the technology of the payments mechanism has been
subject to gradual changes as well as to innovations. These innovations have resulted in rearrangements of payments procedures and related changes in the average stock of money required to accommodate a given flow of payments.
Velocity, however measured, is a statistical averaging of money which moves
with money that is mostly at rest and which is held for liquidity and other purposes
rather than as a means of payment. The demand for money balances for liquidity
purposes will affect velocity independently of the volume of payments for current
output and of financial(assets) transactions. Although demand for liquidity itself
is related, in a complex manner, to the level of economic activity, it is not necessarily related to its current level alone or to the levels of interest rates associated
with such activity. The demand for money, as one form of liquidity, is in turn influenced by anticipation of future requirements and by the relative attractiveness of
competing liquidity instruments. Thus, velocity of money and of demand deposits,
its main component,is influenced by changes in the demand for liquid assets and by
the extent to which such demand is satisfied by holding assets other than money.
The almost continuous rise in velocity after World War II reflects the diminishing
role of money as a means of liquidity and its more efficient use as a means of
payment.
Indeed, changes in technology relevant to the function of money as a means of
payment are paralleled by changes in the options open to holders of money as a
supply of liquidity. These options are continuously affected by changes in the relative attractiveness, in terms of yields offered and risks attached, of various alternative instruments of liquidity or repositories of savings. Such changes are rooted in
10

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

fluctuations in business and credit conditions as well as in the usual institutional
frictions and time lags involved in the spread of innovations. In particular, postWorld War II changes in velocity cannot be understood without taking into
account innovations in instruments and processes, changes in legal and institutional
arrangements, and holder attitudes and preferences.
Thus, one of the reasons for change in the velocity of money over time must be
sought in the whole range of conditions and forces that alter the position of money
in relation to other liquidity instruments and other means of accumulating financial assets, as well as in relation to those forces that influence demand for payments
purposes. Some of the longer run forces, such as the growth of our economy and
the rise in the price level since World War II, affect—though not necessarily in the
same way—both these main demands. But, while demand for money as a financial
asset is best discussed in terms of portfolio balance, the demand for transactions
balances is also influenced by purely institutional and technological factors.
All this has not been ignored. Many writers acknowledge such influences, and
others at least footnote them properly. Published research, however, tends to focus
on the influence of interest rates and income rather than on the payments
mechanism. Interest elasticity of the demand for money has proved to be a more
attractive subject of inquiry than the stability of bank float or the impact of the
lockbox system on the speed of check collection. But since changes in payments
technology and institutional arrangements seem to have played such an important
role in the postwar rise in velocity, we shall review some of the technological factors which have contributed to the increase since World War II.
As velocity reflects shifts in attitudes toward money and liquidity, changes in
payments habits, new cash management and banking techniques, and variations in
the composition of payments flows, as well as other factors, the present study combines a scrutiny of the historical record with an examination of the underlying
institutional and attitudinal factors. It brings up to date discussion of changes in
payments patterns and banking techniques first examined nearly a decade ago.'The
plan and scope of the present monograph differ, however,from those of its predecessor. Some of the descriptive material relating to institutional and technical
factors has been condensed because adequate description is now readily available
elsewhere. On the other hand, it has seemed desirable to draw from the large vol-

3See George Garvy, Deposit Velocity and Its Significance (Federal Reserve Bank of New York,
November 1959).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

ume of relevant published research, including econometric studies of the demand
for money.
The velocity of money can be measured in a variety of ways. This study focuses
on the two that are most widely used, namely, income velocity (V)) and transactions velocity (Vt).Since much of current economic analysis and policy research
centers on levels of aggregate output, it is not surprising that income velocity has
come to play a more prominent role in academic literature than transactions
velocity, which measures the intensity of use of the major component of the money
supply, i.e., private demand deposits. Obviously, when analyzing financial-policy
actions directed at influencing income and employment levels, income velocity is
the more appropriate magnitude. Changes in flows of business and personal payments and in the money-transfer mechanism, on the other hand, can best be traced
through the transactions approach. The two velocity concepts are clearly interrelated, and both ratios are significantly influenced by the degree to which, in the
long run as well as cyclically, assets other than money acquire the attributes of
"moneyness" and substitute for cash balances as liquidity reserves. At times, the
two velocity measures move in opposite directions,for reasons to be discussed below.
The original version of this monograph concentrated on transactions velocity
not only because the Federal Reserve Bank of New York pioneered as early as
1919 in developing this particular measure' but also because it was evident at the
time that important changes in the technology of payments were taking place. These
changes were given little recognition in academic discussion of the demand for
money. They are, moreover, best studied in relation to patterns of aggregate payments rather than to the value of the final product. In the present completely
revised version, the two measures of velocity are given equal attention. Since the
analysis of velocity and that of the demand for cash balances are essentially alternative approaches to the same problem,some of our discussion is cast in terms of the
demand for money.
One justification for paying the attention that we do to transactions velocity lies
in the fact that a major part of the postwar rise in velocity reflects technological
changes in our payments mechanism. Analysis of Vt emphasizes the dependence

4See George Garvy, The Development of Bank Debits and Clearings and Their Use in Economic
Analysis (Washington, D.C.: Board of Governors of the Federal Reserve System, 1952). A revised
version was subsequently published under the title Debits and Clearings Statistics and Their Use
(Board of Governors of the Federal Reserve System, 1959).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

of the demand for money on its technical efficiency in discharging payments obligations. Current discussion of an impending, though not imminent, shift from a
check-ridden to a checkless society illustrates the extent to which the demand for
money, and the velocity of its use, depend on specific technical arrangements.
Since the war years of excess liquidity, velocity has been increasing almost continuously in recessions and prosperity alike, though at differing rates. From a low of
1.97 in 1946, V,increased to 4.59 in 1968;over the same years, V,increased from
13.4 to 36.5. After rising substantially and almost continuously in the postwar
decades, present levels are not much different from those attained in 1929, the
previous peak. A number of important issues have been raised as a result of the
postwar trend in velocity, aside from the perennial question of whether or not it is
approaching some "velocity ceiling". First, there is the question of whether
changes in velocity interfere with the effective execution of monetary policy, since
such variations tend to offset changes in the money supply. A lively debate has
been generated over the extent to which restrictive monetary measures aimed at
reducing the rate of the money supply growth have been hampered or even negated
by the more intensive use of money balances. Other policy-related questions have
also been raised: Do shifts of funds between banks and nonbank financial institutions exert a destabilizing influence? To what extent is the behavior of velocity a
function of interest rates? The list can be extended, but it is clear that an understanding of the reasons for short-term movements of velocity and for longer term
trends is essential to the formulation of monetary policy. Academic debate of
these and similar questions extends to attempts to rationalize postwar velocity
changes in terms of changes in the demand for money. In doing so, the traditional
Keynesian demand motives for holding cash balances have been subject to searching reappraisal, and the long-accepted view regarding the "uniqueness" of commercial banks among financial institutions has been questioned. A number of
specific conclusions have been deduced from such reexamination, including the
position taken by some economists that credit control should replace control of the
money supply as the main objective of monetary policy.
Reasons for holding money balances are discussed in Chapter 2. Chapter 3
examines some of the factors influencing the main payments flows. Problems of
measuring rates of money turnover are discussed in Chapter 4, and the statistical
record is reviewed in Chapter 5. Chapter 6 deals with forces affecting velocity in
the long run as well as over the business cycle. Implications of recent changes are
assessed in the final chapter. In order not to burden the text, some of the technical
detail has been relegated to two appendixes.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

2. The Demand for Money
This study focuses on the narrow definition of money. Our preference for a definition of money limited to currency and demand deposits is rooted in the conviction
that identification of money with the means of payment provides the most satisfactory analytical choice.
Of its various attributes, money plays a central role in the payments process.
In a market economy, money is a carrier of options;1 in fact, demand for money as
a means of liquidity is a derivative of its role as the universal and absolute means of
payment. The use of money as a liquidity reserve is predicated on its immunity to
credit risk and its immediate exchangeability.
Of the several functions performed by money only a few can be dealt with
quantitatively; among them, the demand for transactions purposes is paramount.
Other reasons for holding cash, such as the "speculative" or "precautionary"
motives frequently discussed, are more difficult to quantify and the related demands
may be overlapping rather than additive.
Borrowers normally obtain credit in order to make payments rather than to add
to balances. Borrowed money spent becomes part of the balance of the payee
(business firms, consumers, governmental units, and others) until respent
and/or used to retire bank (or other) debt. Even though extension of bank credit
initially adds to the money supply, the aggregate demand function for cash balances
by all spending units in the economy must not be confused with the demand for
bank credit.
The reasons that individual economic groups wish to hold balances vary, but
clearly the need to support an anticipated flow of payments is the most important.

'In contrast to its role in centrally planned "command economies". See George Garvy, Money,
Banking, and Credit in Eastern Europe (Federal Reserve Bank of New York, September 1966).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

The level of an average adequate balance depends, in addition to the level of
expenditures, on the normal pattern of the flow of funds, the degree of probability
of erratic peaks and troughs in these flows, and the ability of units to replenish their
balances rapidly by borrowing or selling liquid assets. Thus, the management of a
cash position could and does differ widely between industries and consumer groups
and even within industries and consumer groups. Attitudes and practices
are also
diversified according to expectations, past experience, and available financia
l management techniques. A technical factor which influences the demand for cash balances is the compensatory balance requirement—well imbedded in the Americ
an
banking scene—whereby loans are made and certain services performed by
banks
on condition that the customer maintain specified minimum or average
balances.
Another is the "day loan" which enables securities dealers to make during the
day
a large volume of payments from relatively small opening balances.'
In contemporary society, ownership of adequate cash balances is not an absolute
prerequisite for acquiring goods and services or for discharging payments obligations. In fact, the relationship between holdings of money and current spendin
g
runs in both directions. The level of cash balances does not constitute a signific
ant
restraint on spending as long as economic units can replenish such balance
s by
conversion of liquid assets or by borrowing. Furthermore, the relevant
time horizon
for spending units includes not only current but also future and past
incomes. The
accumulated stock of financial assets and the state of indebtedness—largely
a result
of past income and deficit spending history—are also relevant, not
only as a determinant of the unit's willingness to spend, but also in establishing its access
to credit.
Another important fact is that for most economic units the flow of income
in the
immediate future is predictable within fairly narrow limits.
Greater assurance of access to money at times of need further weakens
the link
between cash balances and future spending. Informal or confirm commerc
ed
ial
bank credit lines for business firms and, more recently, overdraf
t facilities for
households ("instant cash" and similar plans) tend to reduce the demand
for cash
in the bank. Issuance of demand claims is profitable to banks
while it is onerous
(in terms of opportunity costs) to the holder. However, to some extent a
bank customer's ready access to additional money balances depends on the level of his
balance, usually in relation to its use. Thus, the holding of balances for
transactions

'See below, page 46.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

the
purposes involves an element (compensating balances) that is also related to
supto
either
future,
the
in
s
resource
cash
ability of the holder to obtain additional
port higher levels of spending (enlarging transactions balances) or to satisfy
demands for asset diversification.
The demand for additional money to spend is not the same as the demand for
money to hold. The demand for bank credit is, in the main, part of the total demand
for credit—command over additional purchasing power; it is not the same as the
demand for money balances as the means of liquidity par excellence which has
been the object of intensive econometric research efforts during the recent decade.
Obviously, as bank lending and investing expand in response to the demand for
additional money to spend and aggregate spending rises, the need for balances to
hold also increases. However, linkages between the demand for bank credit, as
well as cash balances, to real output are complex. In particular, the extent to which
the economy's demand for credit is satisfied by banks rather than other sources
depends on a variety of factors which may or may not be relevant in determining
the demand for money held in relation to the level of output and the demand for
other liquidity instruments.
The relative position of the commercial banking system in the total flow of loanable funds reflects, in each given year, the extent to which the Federal Reserve
System has provided reserves to permit commercial banks to accommodate the
demand for credit beyond the amounts available as a result of the decisions of the
various economic units to spend less than their current income. In war years, and
when for other reasons the budget was in deficit, some or all of additional bank
credit was created for the benefit of the United States Treasury. The importance
of financial intermediaries as a source of credit has also fluctuated in relation to
commercial banking and to direct lending. Changes in the degree of intermediation
of credit through nonbank financial intermediaries are in part reflected in their
demand for cash balances at banks, and at times in loan demand as well, since
financial institutions maintain demand balances with commercial banks to meet
operating needs for cash and, in particular, to provide holders of their liabilities
with the means of payment as required.
While money is created predominantly through the lending and investing activities of banks,the extent to which expansion of bank credit will result in an addition
to the money supply depends on the economy's willingness to retain money so
created. These decisions, including those to convert part of the additional money
than
supply into income-yielding time deposits, are made largely by units other
balances
created
loanas
place,
first
those which cause bank credit to expand in the
16

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

are disbursed to purchase goods and services, to pay taxes, or for other purposes.
The outcome thus depends largely on decisions of units receiving payments, not of
those originating loan demand.
The efficiency with which money is used by various spending units and economic
sectors depends upon a number of factors, some of which are of an economic and
others of a technical and institutional nature. Among the former, the availability
and cost of credit is generally recognized as meriting special attention. Some of the
latter have to do with differences in the timing of receipts and expenditures. Others
can be traced to the space element in the payments process. When payment to a
distant point is made by check,some time will elapse before the check is cleared and
charged against the payer's account. The greater the proportion of payments made
at distant points, the more will turnover rates be affected by delays in receiving
check payments and in collecting the proceeds. The technology used in collecting
checks and making the required bookkeeping entries has an important bearing on
the efficiency with which the money supply is used and thus on the demand for
cash balances, a subject more fully discussed in Chapter 6.
Autonomous changes in transactions levels usually generate accommodating
shifts in the magnitude and composition of money demand between transactions
balances and asset ("liquidity") balances which tend to be reflected by movements of velocity in the same direction as the changes in economic activity. It is, of
course, not easy to ascertain to what extent a rise or fall in velocity is due to a
change in composition of demand deposits, resulting from a shift in the purposes
for which they are held, and to what extent a rise or fall in velocity is due to a more
or less efficient use of deposits in making payments. In addition to purely technical
factors, the demand for money responds to a variety of other factors, some of
which are of a continuing nature, others arising from shifts in expectations and
preferences—the latter frequently related to the whole constellation of interest
rates, transactions costs, and other supply conditions which determine the opportunity cost of money in relation to other liquidity instruments.

DEFINITION OF MONEY

There is no general agreement on the proper definition of the money supply.
Economists disagree as to which specific assets correspond most closely to the
theoretical concept of money. The appropriate or preferable definition of financial
assets constituting money differs from country to country, depending on numerous
factors, such as institutional arrangements and the preferences of the individual

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

categories of transactors.' And, of course, these arrangements and preferences are
not stable over time.
An important consideration concerns whether or not and which time deposits
should be included.' Many monetary analysts exclude savings and other time
deposits from the money supply on the grounds that, while they are (in contrast to
other money substitutes) a commercial bank liability, they must be converted into
checkbook or folding money before they can be spent. Even though time deposits
cannot perform the payments function of money, they can and do,in common with
other money substitutes, satisfy in varying degrees the liquidity needs of at least
certain sectors of the economy. Commercial bank savings deposits, for instance,
compete with savings and loan shares as an outlet for financial savings of households, while negotiable certificates of deposit (CD's) issued by commercial banks
compete with other money market instruments as a means of fulfilling the liquidity
needs of corporations.
The question of the degree of "moneyness"of savings and time deposits' has been
subject to a lively debate, especially since the publication of the important studies
by Gurley-Shaw and Friedman-Schwartz.' Various statistical investigations suggest

'Monetary authorities of various countries, no less than economists, have tried to come to grips
with this problem when publishing monetary statistics. See, for instance, the Monetary Survey of the
International Monetary Fund. Once-and-for-all decisions are difficult to make. Thus, particularly since
the late 1950's, the rising velocity of circulation of certain savings balances, evidencing their greater
use in making payments, has caused the Netherlands Bank to classify certain types of savings balances
as "near moneys".
4Proponents of the view that changes in the money supply are the direct cause of change in the
level of economic activity do not agree on the proper definition of the money supply. Milton Friedman's empirical tests are based on a definition which includes time deposits of commercial banks,
while Leona11 C. Andersen and his associates at the Federal Reserve Bank of St. Louis prefer the
conventional narrow concept.
'According to the call report of June 1968, 62 percent of all time deposits in commercial banks are
personal savings deposits; the remainder are business, government, foreign, or bank time deposits.
Inquiries conducted by the Federal Reserve System in the midthirties, together with data currently
collected by the American Bankers Association and the Federal Reserve Bank of Chicago, indicate
that withdrawals from savings accounts during a given year amount to about half of the average
balance. In other words, savings deposits turn over about once every two years. This ratio has been
quite stable from year to year (except in 1966 when it rose significantly) and appears to be little
influenced by changes in business activity. For a detailed discussion, see George Garvy, "The Velocity
of Time Deposits", Journal of the American Statistical Association (June 1953). For a more recent
discussion, see "Slowing in Savings Deposit Growth and Turnover", Business Conditions (Federal
Reserve Bank of Chicago, June 1964) and "The Rise in CD's at District Banks", ibid.(October 1965).
6John G. Gurley and Edward S. Shaw, Money in a Theory of Finance (Washington, D.C. 1960)
'
and Milton Friedman and Anna J. Schwartz, A Monetary History of the United States, 1867-1960
(Princeton, New Jersey, 1963). See also David M. Jones, The Demand for Money: A Review of the
Empirical Literature (Federal Reserve Bank of New York, 1965, mimeographed), briefly summarized
in the Federal Reserve Bulletin (February 1966), pages 164-65.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

that marginal changes in the flow of income are related to marginal changes in the
stock of savings deposits as well as to changes in the stock of money, narrowly
defined. One investigator, using data for the period 1947-66, found that the stock
of savings deposits, variously defined, possessed some significant degrees of moneyness which decreased as the concept of savings deposits gradually broadened to
include savings deposits at mutual savings banks, postal savings deposits, and ultimately savings and loan shares along with time deposits at commercial banks.'
Tests along these or similar lines merely indicate that, in some periods at least,
marginal changes of income are correlated in varying degrees with changes in the
stock of selected financial assets. They do not establish (or disprove) the moneyness of time deposits or other near moneys. Instead they merely suggest that
changes in financial assets are among the factors which, in addition to money,
influence the level of economic activity. And, to use Professor B. Pesek's words,
"to test, first, all sorts of conglomerates of liquid assets and only subsequently give
the name 'money' to one that correlates best makes monetary theory vacuous since
it is not subject to contradiction; surely a modern computer will always come up
with some conglomerate that correlates well with income"!
The interaction of a rising opportunity cost of money and of advances in cashsaving technology has induced consumers, no less than other major categories
of economic units, to rely on money substitutes for meeting a large part of their
liquidity needs, converting such claims into money as needed. For all practical
purposes, savings deposits at mutual savings banks and savings and loan shares are
regarded by their owners as adequate substitutes for commercial bank savings
deposits. Thus, the proper dividing line is not between total commercial bank deposits and other money substitutes, but between demand deposits and all other

'G. S. Laumas, "The Degree of Moneyness of Savings Deposits", American Economic Review,
(June 1968), pages 501-3. In fact, by including all commercial bank time deposits, this analysis is
not limited to savings deposits, but also includes negotiable certificates of deposit and other time
deposits held by corporations and others. See also T. H. Lee, "Substitutability of Nonbank Intermediary Liabilities for Money: The Empirical Evidence", Journal of Finance (September 1966),
pages 441-57.
Other investigators, who limited their analysis to commercial bank time deposits, arrived at similar
conclusions (and a virtually identical estimate of moneyness)for the eleven years (1929-39) preceding
World War II and for the years following it, but found that such deposits possessed a negative degree
of moneyness for other periods. R. H. Timberlake and J. Forton, "Time Deposits in the Definition
of Money", American Economic Review (March 1967), pages 190-94.
8The Journal of Finance (December 1968), page 904.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

deposit claims (including savings and loan shares).9 How households choose to
distribute their assets among the various forms of liquidity (money, savings deposits, or in other ways) depends upon a number of considerations, including relative rates of return, transactions costs, safety, and convenience. Instead of searching
for a "broad" definition of money that would be consistently valid over a fairly
protracted period of time—say one generation—it would appear preferable to focus
our investigation on the narrower definition of money and to explore how changes
in asset preferences have influenced changes in its velocity.

MONEY SUBSTITUTES

Money is different from other financial assets because it does not yield any explicit
income!'Between money and largely illiquid assets, such as pension rights, stands
an array of financial claims, each unique in one or more respects and each at the
same time competing for inclusion in portfolios of liquid assets. The degree of
liquidity of money substitutes (also referred to as "near money") is usually defined
in terms of convertibility into money, the liquidity instrument par excellence.
Most but not all money substitutes are liabilities of financial intermediaries:
short-term Government securities and commercial paper issued by industrial corporations,for example, are not. On the other hand, of the two nondepository financial intermediaries with the largest volume of assets—life insurance companies and
personal trust funds, including pension funds—only the first provides a limited
degree of liquidity to individuals (through policy loans); neither supplies any
liquidity instruments that can be used as money substitutes.
While money substitutes do not possess liquidity to the same extent as money,
they do offer the benefit of an interest return. Still more important, this return
comes with no great sacrifice of liquidity, for claims against thrift institutions and
money market instruments are readily convertible into cash at the option of the

9See Michael J. Hamburger, "Household Demand for Financial Assets", Econometrica (January
1968). See, in particular, his conclusion on page 105 where references are given to several studies
confirming as well as disagreeing with this conclusion. This article also contains an extensive bibliography on the demand for money. See also George G. Kaufman, "More on an Empirical Definition
of Money", American Economic Review (March 1969).
10This is not a universal characteristic of money. The prohibition against payment of interest on
demand deposits did not exist in the United States prior to 1933, and interest is paid on sight deposits
in some foreign countries.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

holder. Money market instruments, other than those issued by the Federal Government, are subject to some degree of credit risk. However, the short-term nature of
the claims and the high financial standing of their issuers reduce loss expectations
to a minimum.
Over the postwar period, money substitutes have come to comprise a steadily
increasing proportion of liquid assets portfolios (see Chart 1), in part as a result
of the upward trend of interest rates. As a consequence, especially since the early
fifties, economic units have been adjusting their liquidity reserves to include a
greater share of interest-bearing claims and a smaller share of currency and

Chart 1. MONEY AS A PERCENTAGE OF TOTAL LIQUID ASSET HOLDINGS
1953-68
Percent'

...„„.....Zonfinancial corporations*
60

30
Householdst
20

0
1953

I
1954

II
1955 1956

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

111111
1957 1958 1959 1960
1961
1962

1963

I
1964

I
1965

I
1966

I
1967

1968

* In addition to money, the liquid assets held by nonfinancial
corporations include time deposits, short- term United States
Government securities, and commercial paper.
t In addition to demand deposits and currency (i.e., money), liquid
assets held by the household sector include time and savings
deposits at commercial banks and savings institutions, short-term
United States Government securities, and savings bonds.
Source: Board of Governors of the Federal Reserve System, Federal
Reserve Bulletin, Flow of Funds Accounts.

demand deposits. While the demand for transactions purposes may be presumed
to be less interest elastic than that for asset-holding purposes, significant increases
in interest rates tend to lead to reassessment of minimum transactions balance
needs.
In the last two decades, numerous attempts have been made to resolve the issue
by econometric analysis of the behavior of various types of financial assets in order
to determine which of them come closest to money, narrowly defined. Various
authors either have widened the definition of money on the basis of their interpretation of the results of their inquiries or have suggested alternatively that certain
groups of assets be considered as money substitutes. Of course, there is no more
agreement as to the empirical scope and theoretical interpretation of such categories than there is on the proper definition of money. The general nature of the
process which, since the revival of monetary policy in the early 1950's, has led to
a massive shift of funds from demand deposits into money substitutes is, however,
fairly simple to describe.
Business firms, households, state and local governments, and eleemosynary
institutions, as well as foreign holders of dollar balances, can reduce the opportunity cost of holding adequate cash balances by investing in money market instruments that are readily convertible into cash and which involve a minimum of risk.
Since the fifties, there has been a proliferation of tailor-made market instruments
and arrangements to meet various corporate needs (as well as similar requirements
of other large depositors, such as municipal governments) and liberalization of time
contracts available to individuals. As shifts from cash into income-yielding liquidity
instruments and back into cash involve certain transactions costs, it became unprofitable in periods of very low money market rates, which characterized the
thirties and the subsequent World War II period, to economize on cash balances.
The situation changed again in 1951, when rates began to rise after the "Accord"
and as various borrowers and intermediaries in the money market began to develop
instruments to meet the specific needs of various groups of demanders for money
substitutes. In recent years, a practice has developed where banks agree to "borrow" from corporate and other large depositors all demand balances above a
certain agreed minimum amount. Since such borrowing—even for one day—is not
prohibited so long as it is secured, redundant balances are absorbed by the bank
and the customer's balance is kept at a workable minimum, showing a higher
velocity than would otherwise be the case.
Periods of high interest rates usually coincide with periods of active and rising
business. In such periods, there is an opportunity to make additional cash outlays
22

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

on goods and services without correspondingly adding to cash balances, and it
becomes more profitable to exchange money for other financial claims. As a result,
measured turnover rates increase. Parallel,, but opposite, reasoning suggests that
money turnover rates decline in periods of low interest rates. It is quite obvious
that shifts in the composition of portfolios of various holder groups between money
and all other liquid assets combined affect cyclical and, more importantly, secular
velocity.

THE ROLE OF FINANCIAL MARKETS

Liquidity is synonymous with salability or shiftability, which, in turn, requires the
existence of efficient financial markets."
The development of the money market, in which the various categories of financial intermediaries participate (although to varying degrees), along with the growth
of large manufacturing and other corporations, has increased the moneyness of
those financial instruments and arrangements that have come to compete with
demand deposits as a means of liquidity. Essential elements in the creation of this
market have been, first, the existence of a large number of participants, with seasonally varying needs and different degrees of sensitivity to cyclical influences and,
second, the availability of a wide range of instruments of differing maturities and
legal characteristics suited to the varied requirements of lenders and borrowers. In
this market, liquidity is provided through shiftability and partly because of the
variety of needs of the participants, as some investors are always willing to trade
maturity for income. This shiftability, in turn, depends to a large extent on the
efficiency of dealers, brokers, and other intermediaries who establish markets, seek
out investable funds, and periodically develop new techniques, such as repurchase
agreements on Government securities. It is the development of the money market,
with the perfection of its operating techniques, and the participation of the banking
system in this market that have been instrumental in narrowing the gap between
money and money substitutes.
While individual units can easily convert money market assets in the market, all

"In recent decades several advanced countries have experienced a very
growth of financial
intermediaries, in some cases paired with a spectacular growth of public rapid
debt, without showing a
parallel substitution of near moneys for money. One reason is that they did not
succeed in developing
a sufficiently broad money market with a proper range of instruments and techniques.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

are
economic units combined can "shift into cash" only to the extent that banks
addito
able and willing to acquire such assets on balance. Thus,the ease of access
tional money, through borrowing or conversion of assets, is influenced by current
and prospective monetary policy.

ITURES
SYNCHRONIZATION OF RECEIPTS AND EXPEND

Money as a means of payment is essential for completing most, if not all, transactions arising in a modern economy. The flow of goods through the various stages
of production and distribution involves payments to the various agents of production, such as labor and capital. Normally, whenever the ownership of either interl
mediate or final products passes from one economic unit to another, additiona
)
involved
debits to the account of one unit or more (when brokers or agencies are
arise. As the flow of revenues is not perfectly matched to the flow of payments,
business units hold cash balances to meet payments when they fall due. The
amount of cash held in relation to the volume of payments depends, in part, on the
opportunity cost of holding money as well as on purely technical factors, such as
the frequency of receipts (which itself depends on billing practices and the characteristic seasonal pattern of sales in each given industry) and the schedule of
disbursements (in which the frequency of wage and salary payments and of purchases of raw materials are among the more important influences). This holds true
for all categories of transactions—business, consumers, nonprofit institutions, and
units at all levels of government.
Mathematical models have been developed to show that the need for cash balances declines with improvements in synchronization of the flow of personal and
business payments to receipts; the shorter the pay period, the smaller the balance
which the average consumer will have to hold in relation to expenditures." The various relevant factors are usually discussed in textbooks under such headings as
"income periods", "overlapping of payments", or "time intervals" between interfirm payments."

income evenly over that period.
"Suppose a worker is paid every four weeks and spends his entire
balance of $200 and the annual turnover rate
With a paycheck of $400, he will keep an average cash
a weekly paycheck: his
is 26 times the average cash balance. But assume instead that he receives
to 104.
average cash balance now is $50, while the rate of turnover increases
York, 1911); J. W. Angell,
"See, in particular, Irving Fisher, Purchasing Power of Money (New
Journal of Economics (February
"The Components of the Circular Velocity of Money",The Quarterly
ibid., May 1938.
1937), and H. S. Ellis, "Some Fundamentals in the Theory of Velocity",

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Changes in typical payment periods occur from time to time, particularly in the
business sphere. Business firms may shift to new payments or billing procedures
(such as cycle billing) with the specific purpose of improving the synchronization
of their cash flows and reducing transactions balances besides spreading out the
work load. Corporate treasurers of large firms with diversified products are usually
better able to anticipate and synchronize payments flows than independent units
linked by market relationships, with all the attendant uncertainties and needs for
protective margins. The fewer the number of independent business units involved
as a product moves through the various stages of production and distribution, the
fewer the number of potential gaps between receipts and expenditures. The precise
way in which growing integration in production and distribution, particularly
through corporate mergers, reduces the need for transactions balances depends on
how payments between formerly separate establishments are handled after they
are combined.
The problem of synchronization is not limited to production and distribution of
current output. It is pertinent also to trading in financial assets, since not all sales of
financial assets are immediately matched by purchases. Transactions in financial
assets, being determined by such considerations as the relative yield patterns of
various types of securities and the liquidity needs of business firms and individuals,
lack the regularity of wage, rent, tax, dividend, and other payments which flow
from the process of producing and distributing current output. As a result, the level
of cash balances arising from, or held in connection with, trading in financial markets varies with business conditions, interest rates, and other factors. In particular,
when investors shift from securities (and other financial assets) into cash, the
velocity of cash balances is reduced.

FACTORS TENDING TO REDUCE THE DEMAND
FOR TRANSACTIONS MONEY

Even though money is first of all a means of payment, not all income-producing or
asset transactions require its use. Numerous institutional arrangements have been
developed to minimize the amount of cash used in trading financial assets. Payments also can be offset or obviated, reducing the demand for cash balances, as
discussed in more detail in Chapter 6. The use of book credit and the offsetting of
interbusiness payments tend to reduce cash needs as well,but itis uncertain whether
the relative importance of industries in which interbusiness financing and/or the
use of book credit are significant has increased or decreased over the years.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Payments may also be made without the use of money, depending on the convenience to the payer and the willingness of the payee to accept (or even to require)
substitutes as well as on habits and facilities offered. For instance, a not inconsiderable volume of goods and services is paid for by endorsing checks received. Credit
instruments circulating as money substitutes are limited now almost entirely to
drafts, but in earlier periods, and even currently in some foreign countries, trade
bills, acceptances, and similar instruments have to be considered as substitutes for
money. Government securities issued for the specific purpose of being accepted
in lieu of cash (tax anticipation bills) and Treasury securities acceptable in payment of estate taxes are other examples of payments not requiring cash.
Finally, various means of payment for special purposes have been developed in
exchange for money. They include traveler's checks, money orders, and even
tokens, such as those used by public transportation systems." Conversion of universal money into special payments instruments increases income velocity by
reducing average money balances held and also affects transactions velocity. It is
unlikely that the proportion of payments made without the use of money is subject
to significant short-run fluctuations (nor is it particularly sensitive to changes in
monetary policy), but it may need to be considered in long-run comparisons.
Important changes in the demand for money stem from the ability of various
categories of holders to increase the efficiency of transactions balances. Specific
policies to achieve such economies are discussed in Chapter 6. More efficient use
of deposit balances is reflected in increased transactions velocity, unless offset by
an increase of balances held for liquidity purposes or for other reasons not directly
related to the volume of current check payments.
Greater efficiency in the use of deposit balances has clearly been one of the main
influences behind the post-World War II rise in the income velocity of money.
Another important influence has been the growing tendency for liquidity and other
reserves and temporarily redundant funds to be invested in money market instruments; fewer and fewer households are now content to let savings accumulate in
nonearning demand deposits.

"For example, the demand for dimes dropped abruptly in New York when subway fares went up
and tokens were substituted for dimes.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

ACTIVATION OF IDLE BALANCES

In older literature, balances not needed for transactions purposes were sometimes bracketed as "idle" or "excess" balances. The distinction between "active"
and "idle" balances is, however, essentially an expository device. Demand deposit
accounts that are virtually dormant for prolonged periods of time (or that are
occasionally increased but not drawn against) are exceptions; nearly all accounts
show some degree of activity. In effect, idle balances are an abstraction referring
to amounts in excess of those normally required to meet a given flow of
payments."
For each business firm, individual, or other spending unit, the amount of idle
balances depends on a number of factors, in addition to interest rates, such as
liquidity needs arising from the nature of the account owner's business and his plans
for the future, the geographic and time patterns of payments, and the requirements
of the banks as to minimum or compensating balances. These factors may be subject to long-run as well as cyclical influences. For example, those who hold cash
for
speculative purposes may change their expectations regarding prices of goods
and/or services as well as yields on securities, and they may decide to use
the
balances they have built up to acquire other goods, services, or securities as the
favorable situations which they have anticipated begin to materialize.
This process, referred to as "activation of idle balances", is best looked upon as
a reshuffling of liquid-asset holdings when interest rates rise. As the volume of payments expands with business activity, some existing balances are found to
be
ample enough to support additional transactions and their turnover rates accordingly increase. Other accounts which are close to minimum needs have to be built
up by borrowing or by converting liquid assets into cash;their rates of turnover may
not change much, as balances and transactions rise more or less in step, but the
growth of such balances will enlarge the share of more active accounts in total
demand deposits. Statistical data recording rising turnover velocity in periods when
the pace of economic activity accelerates reflect additional debits generated by
various cash management activities as well as a cyclical reduction of the share of
idle money in the total money supply.

"For a pioneering study of "excess cash" estimated by applying minimum
cash-sales ratios
achieved in 1929, see Avram Kisselgoff, "Liquidity Preference of Large Manufactu
ring Corporations", Econometrica (October 1945), pages 334-44.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

COMPENSATING BALANCES

It is usually assumed that the size of the average cash balance held by each economic unit is regulated by its transactions and liquidity requirements. In fact, however, balances also serve to compensate banks for the services they perform,including the handling of payments and the providing of credit facilities. Indeed, a substantial part of banking services, in particular the maintaining of regular business
accounts, is paid for in the United States by appropriate compensating balances.
Such balances are a large proportion of total demand deposits, although the total
amount of deposits held as compensatory balances is not known." This practice is
peculiar to American banking, Canada being the only other country in which compensating balances are used. Furthermore, the amount of compensating balances
lodged with the commercial banking system and the changes in this amount
caused by cyclical or other influences are significant factors in explaining the behavior of velocity.
Individuals as well as businesses usually have, within certain limits, an option
either to pay a service charge for the handling of their accounts or to compensate
their banks indirectly by maintaining balances on which the banks can earn an
income equivalent to the service charge." The laws often make it advantageous for
individuals to pay for banking services through compensating balances. The
depositor forgoes income on the balance maintained, but in many cases he would
have had to invest a larger amount in order to earn enough income after taxes to
pay bank charges.
Corporations pay service charges for routine bank services (such as account
activity, preparation of payrolls, wire transfer of funds, credit inquiries, etc.) only
in exceptional cases, for instance, when services are used intermittently or when
the bank's deposit requirements are judged by the corporation to be excessive.

Monthly Review
,6See also Jack M. Guttentag and Richard G. Davis, "Compensating Balances",
also Thomas Mayer and
(Federal Reserve Bank of New York, December 1961), pages 205-10. See
n", The National Banking
Ira 0. Scott, Jr, "Compensatory Balances: A Suggested Interpretatioconducted
by them suggestReview (December 1963), pages 157-66, which cites the result of a survey and that most banks admining that banks, in fact, pay an imputed interest on compensating balances
"Compensating
Gibson,
E.
ister the compensating balance requirements in a flexible manner, and W.a recent sample survey, see
Balance Requirements", The National Banking Review (March 1965). Fors,
the Financial
H. G. Hamel and F. J. Walsh, Jr., Commercial Banking Arrangement Managing
Function, No. 3(New York: National Industrial Conference Board, 1968).
by keepnThe United States Government also compensates commercial banks for certain services Treasury
ing appropriate balances; see H. J. Cooke,"Managing the Treasury's Cash Balances", in The
and the Money Market (Federal Reserve Bank of New York, May 1954) pages 7-11.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Those bank services to business firms that normally give rise to the payment
of
explicit service charges are corporate trust services (e.g., the issuance of dividend
checks and stock certificate transfers). In recent years, the institution by banks of
various "bookkeeping"-type services has led to an increase in the amount of bank
charges.
Although the idea of service charges began to take root before the turn of the
century, by 1929(when banks still paid interest on demand deposit account
s)
only a little more than one third of all banks had adopted the practice of charging
for services rendered. The depression, but also the discussion of banking
codes
under the National Recovery Administration, boosted the movement toward the
application of service charges. Subsequently, account analysis became almost
general practice, and payment for banking services by maintaining adequate balances was encouraged." In general, since World War II, and particularly in recent
years, there has been a tendency to place the "pricing" of banking services on the
same basis as in nonbanking businesses." Charges levied for various types of services,and the rate used to compute earnings credit,vary considerably among banks.
Banks periodically analyze the activity of business and other large accounts in order
to determine their profitability. Activity charges are computed through a fairly
standardized analysis based on the principle that earnings credit allowed on average
(or minimum) balances, after deduction for applicable legal reserves, should
be
equal to the actual cost incurred plus a fair profit margin. Most banks compute
earnings credit on the basis of collected balances, but others compute them on
ledger balances.
Corporate treasurers usually review independently the account analyses made
by their banks. In considering the proper level of compensating balances, they may
allow for perhaps an even wider range of services than those considered by the
banks by taking into account various intangible services performed by banks.
In
addition to the routine banking services, services rendered to the customer's employees, such as assisting transferred employees in locating and financing homes,

"The American Bankers Association issued in 1935 its first Manual for Determini
ng Per Item Costs,
followed in 1939 by Uniform Account Analysis.
"See James P. Furness and Paul S. Nadler,"Should Banks Reprice
Services?" in Harvard
Business Review (May-June 1966). Whether and how profitable theCorporate
ment is for the commercial banks is a matter of debate. See Richard compensating balance requireG.
Davis
and
Jack M. Guttentag,
"Balances Requirements and Deposit Competition", The Journal
1963) and William G. DeWald,"The Economics of Compensating of Political Economy (December
Balances", Iowa Business Review
(July 1965), and the literature quoted therein.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

goodwill or
may also be given recognition. Considerations relating to community
a role.
play
also
the maintenance of long-established banking connections may
balance with
Some corporations maintain as a matter of policy a certain minimum
ories."
each bank in their network of local deposit
sually 15
In general, compensating balances are determined as a percentage—u
about 10 perpercent to 20 percent—of the amount of borrowing outstanding, or
er nor mortgage
cent to 20 percent of credit lines granted. Normally,neither consum
al loans such as
financi
of
types
loans involve commensurate balances, nor do most
deposits of a
total
those to brokers. The share of commensurate balances in the
This is significant
given bank will depend on the composition of its loan portfolio.
ies and where
when comparing turnover ratios among individual banks or localit
s in the
change
of
light
the
in
assessing long-run changes in transactions velocity
composition of assets of the banking system as a whole.
determined by the
The total amount of compensating balances maintained is
, to compensate
primary functions which they are intended to perform, namely
to assure adeand
s
banks for costs incurred in rendering a wide range of service
satisfy part or all
quate access to credit. Yet, at the same time, such balances may
balances are
of the liquidity requirements of their owners. Since compensating
be drawn
can
they
es,
balanc
um
normally based on average rather than on minim
of incurcost
at the
down temporarily to meet unexpected drains on funds, usually
. Many business
ring service charges, and rebuilt when the customer is "in funds"
their liquidity reserves.
customers rightly consider compensating balances as part of
or expect their
Banks, in particular large institutions, also ordinarily require
extension of credit
customers to maintain appropriate balances in return for the
balances, although occalines." Such balances are also referred to as compensating

Banks (Ann Arbor, Michigan: 1957), Chapter 6. Also
"See George Katona, Business Looks at
Spending", Journal of Finance (December 1953), pages
J. E. Walter, "Liquidity and Corporate
369-87.
two fifths of commercial banks with total deposits
21According to a recent sample survey, roughly
ating balances, while for larger banks, with deposits of
of less than $20 million require compens
to nearly 100. Even in the case of relatively small banks
$150 million and over, the percentage rose
ge is about 85. The use of compensating balance
percenta
the
million),
$50
and
million
(between $20
ead since the emergence of greater rate fluctuarequirements has become increasingly more widesprof all commercial banks requiring compensating
half
tions after the "Accord" of 1951. Only about
practice was limited mostly to larger banks. Thus,
balances in 1964 did so before 1951, when thedeposits
who required compensating balances in 1964
in
billion
$1
over
with
banks
80 percent of the
with deposits of under $20 million. Nevins D.
banks
the
of
percent
15
only
but
1951,
before
so
did
Balance Requirements: The Result of a Survey",
Baxter and Harold T. Shapiro, "Compensating"Credit Lines and Minimum Balance Requirements",
Journal of Finance (September 1964). See also573-79 for data based on the October 1955 survey of
Federal Reserve Bulletin (June 1956), pages

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sionally "commensurate" or "supporting" balances is used to distinguish them from
balances required to compensate banks for checking and similar services. Commensurate balances apply with varying frequency to almost all types of business
borrowers, and are determined by the category and size of the customer. Some
banks require commensurate balances only from certain categories of borrowers,
such as finance companies.
Frequently, corporations which do not establish formal credit lines maintain
sufficiently large balances with their principal banks in order to facilitate access to
credit in the case of urgent need. Indeed, most banks that have no formal minimum
balance requirements would take a customer's usual deposit balance into account
when providing loan accommodation and in setting the interest rate on a loan.
Banks look at commensurate balances as an important aspect of overall bankcustomer relationships. Compensating balances are sensitive to interest rates but
only to a limited extent. The earnings credit, on the basis of which the size of balances required to compensate for account activity is computed, moves with the
general level of interest rates, but rates used for computing earnings credit are
adjusted only when major changes in interest levels take place, and even then with a
lag." Commensurate balances to support borrowing and related goodwill balances
are, on the whole, also rather insensitive to fluctuations in interest rates. Nevertheless, under the pressure of high or rising interest rates, some depositors maintaining
a large number of accounts will tend to eliminate certain low-activity and duplicating accounts. On the other hand, bankers will place more emphasis on balances
when loan demand is strong and use commensurate balances as a credit-rationing
device.
The lodgment in the deposit structure of a large volume of compensating balances therefore has the effect of dampening cyclical swings in velocity. Their aggregate volume tends to be related to account activity and loan volume; and any
increase in account activity and loans granted or committed calls for an increase in
compensating balances. Compensating balance requirements may be raised, their

21(continued from page 30):
commercial and industrial loans at member banks, and D. P. Jacobs, "Sources and Costs of
Large Sales Finance Companies", Consumer Instalment Credit, Part II, Vol. I (WashingtoFunds of
n, D.C.:
Board of Governors of the Federal Reserve System, 1957), pages 341-52.
221f, in a period of expansion, reserve requirements are raised, higher earnings credit
may be ofiset,
in part or entirely, since required reserves are deducted from compensating
balances on which the
earnings credit is based.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

wers, and perhaps most imporuse may be extended to a greater number of borro
enforced. To the extent comtant the requirements are definitely more vigorously
ces that corporations would volunmensurate balances come to exceed the balan
At the same time, the fact that
tarily keep, the effective yield on loans is increased.
deposits) is reduced acts as a brake
the total volume of usable bank credit(demand
nsating balances held for various
on velocity: the greater the amount of compe
is the computed velocity. The relapurposes in total demand deposits, the smaller
economic activity and the level of
tionship between compensating balances and
specific recognition in econometric
interest rates is sufficiently complex to deserve
when analysis is extended to the
studies of the "demand for money", especially
ally disregarded this relationship."
twenties; unfortunately, such studies have gener

MONEY
MODELS OF DEMAND FOR

nt basis has encouraged conAvailability of a variety of pertinent data on a curre
ls of the demand for money.
struction of increasingly complex econometric mode
zed in numerous ways. Simple
The relationship of money to spending can be analy
actions demand for money, and
models usually take as a starting point the trans
on near moneys tends to be
they demonstrate how the level of returns available
associated with efforts to economize on cash balances.
. Since direct and independent
Transactions balances are not reported separately
or the demand for financial
measurements of either the aggregate payments flow
the demand for money relate
assets are also not available, empirical studies of
omic activity. Among these,
transactions demand to some index of current econ
rs figures predominantly because
gross national product (GNP) in current dolla
various velocity series has been
of its comprehensiveness. The time profile of the
bles of a cyclical kind (e.g., interest
explained in terms of the interaction of varia
average income or accumulation of
rates) or of a long-run nature (such as rising
ns in structural changes in the
financial assets). Others have sought explanatio
page 58).
economy or in payments streams (see below
l, which concludes that the transmode
t
semen
isbur
The simple receipt-and-d

and holding of comgenerally paid interest on demand deposits,major
reasons why in
23Prior to 1933 commercial banks
was not widespread. This is one of the
sense
t
curren
the
in
es
balanc
in the twenties,
ing
than
pensat
lower
erably
consid
was
ts
deposi
d
deman
of
y
the fifties the transactions velocit had been made in collecting and managing funds by large holders
in spite of the great progress that
of demand balances.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

actions balance will average half of the total flow (assuming that cash is available
at the beginning of the period and that payments have a fair degree of regularity),
can be made more realistic by allowing for investment of part of the balance in
income-earning assets and resorting to short-term borrowing to meet scheduled
disbursements during the latter part of the period. Such a model can be made even
more realistic by explicitly introducing the cost of engaging in asset-switching
operations and the cost of negotiating loans. Tested against empirical data for the
period 1947 to 1965, one such model yields significant results which show that the
average transactions balances held in the form of cash are related positively to the
cost of borrowed funds and negatively to the return on long-term investments."
Demand-for-money models can be, and have been, made more complex by introducing the diversification demand for money. The demand for money as a financial
asset is not related importantly to the current level of economic activity. The volume of financial assets existing at any given time is the cumulative result of experience over a period of years, and thus diversification demand for money is not
directly related to current product or income alone. Neither is it necessary to
assume that lags involved in the demand for money in relation to the various determining factors are stable over time or identical for all sectors.
Numerous empirical studies have been undertaken to clarify the question of
whether the demand for money is primarily a function of money's unique characteristic as a universal means of payment (and thus related to some variant of
income) or a function of wealth as its owners attempt to diversify holdings (thus
making portfolio diversification an important factor of demand), or both. Virtually all such studies have found money balances to be responsive to money rates,
but there is little agreement on whether the cost of bank borrowing or rates on assets
competing with money and whether short- or long-term rates are relevant. Since
business expansions are normally accompanied by rising interest rates, it is not
surprising that many researchers have found in fluctuations of interest rates a ready
explanation for the cyclical behavior of rates of money turnover." Though statistically significant relationships between interest rates and either of the two measures

"E. L. Whalen, "An Extension of the Baumol-Tobin Approach to the Transactions Demand for
Cash", Journal of Finance (March 1968).
25Since the publication of the frequently quoted article by Henry A. Latane, "Cash Balances and
the Interest Rate: A Pragmatic Approach", Review of Economics and Statistics (November 1954),
pages 456-60, the literature on this subject has grown extensively. Widening the concept of money
usually results in reducing interest elasticity.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

of velocity are readily established, it is not possible to determine precisely
to what extent fluctuations in velocity are caused directly and uniquely by changes
in the attractiveness of yields available on money market instruments (or thrift
deposits)."
Demand-for-money models can be complex. They may, for instance, allow for
a "learning process" with regard to the demand for money substitutes. They can
recognize the fact that costs incurred in establishing a money-management function
within a corporate structure are not simply written off when rates drop below a
certain level. Money market participation is continued in such periods in order,
among other things, to preserve access to the market at times when participation
in it would prove profitable. This "institutionalization" of response patterns, operating independently of initial economic stimuli, has not as yet been adequately explored.
Demand-for-money models as a rule make a number of simplifying explicit or
implicit assumptions, many of which are questionable, such as the assumptions
that cash holdings are proportional to outlays and that lags are constant over time.
The general difficulty with all these models is the assumption that motivation can
be dealt with on a universal basis, using national (aggregate) series for empirical
verification, while in fact relevant factors underlying the demand for money differ
significantly among major economic segments and spender categories." The demand function for money must be differentiated for the various categories of
holders of cash balances. Patterns of financial asset holdings vary widely. Cash
inventory demand of individuals differs from that of business.
Much of the underlying theoretical literature is cast in terms of the demand of
individuals for money balances. In fact, business firms are the largest single holder
group, and their cash balances are more likely to be subject to cyclical and long-

"For a theoretical model of cyclical variations in velocity that does not consider interest rates
at all, see Milton Friedman, The Demand for Money: Some Theoretical and Empirical Results, Occasional Paper 68 (New York: National Bureau of Economic Research, 1959). Karl Brunner and Allan
H. Meltzer—"Economies of Scale in Cash Balanees Reconsidered", The Quarterly Journal of Economics (August 1967), pages 422-36—deny that economies of scale contradict the quantity theory of
money and question the claim that such economies in the demand of business firms for money are
implicit in the Baumol and Tobin models.
27For one of the few exceptions, see M. I. Hamburger, "The Demand for Money by Households,
Money Substitutes, and Monetary Policy", The Journal of Political Economy (December 1966).
Unfortunately, data on money holdings by the main economic sectors are not available, and even
the limited data on deposit ownership for 1943 to 1961 and estimates prepared in connection with the
flow of funds are based on a set of broad assumptions, usually held unvaried over long periods of time,
thus failing to reflect many important structural and institutional changes as well as changes in
processes and procedures.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

run changes than those held by consumers. Thus, while household wealth may be
a factor in the diversification of money, it is not pertinent for cash holdings of
corporations which are estimated to hold half of the money supply. Rates on negotiable CD's have been relevant since 1961 in determining corporate demand for
cash, but they are significant for only a comparatively small group of wealthy individuals. Rates available on savings and loan shares are probably more important
for individuals as a factor in determining the attractiveness of demand as well as
of time deposits at commercial banks, but are largely irrelevant for corporations
which cannot hold such shares.
Emphasis on interest rates and/or on the income or wealth variable (e.g., GNP,
Friedman's "permanent income") has the limitations inherent in all types of analysis searching for a universally valid "first cause". Yet, to explain money demand
in the middle sixties, for instance, as responding in an unvarying way to interest
rate movements is wholly to ignore the facts that liquidity management has become
an integral part of the overall financial management function and that economizing
on balances has become a structuralized part of portfolio management. Looked at
from the corporate treasurer's point of view, money has fallen in stature from
liquidity par excellence to one of several liquidity instruments(that is, as a liquid
asset having a number of substitutes), to a costly inventory, the amount of which
should be reduced at every opportunity.
Before proceeding further, we should emphasize that the available statistical
data on the money supply, including its short-term changes and ownership by
principal economic sectors, are quite deficient, and the prospects for remedying
them in the near future are dim.Indeed, empirical studies of the demand for money
are beset by a variety of theoretical and technical problems. One important difficulty is that adequate empirical data has become more readily available only since
the end of World War II, a period of rapidly rising income and mild cyclical fluctuations in the United States and elsewhere. As a result, any time series is dominated by a strong upward trend; but the problem of the high degree of correlation
between all relevant series extends to their similar response to cyclical influences
as well. Much of the published research related to the use of money, whether
approached in terms of demand or velocity,is handicapped by these limitations. One
conclusion emerges clearly from the historical record whatever the shortcomings of
the underlying statistical series: the velocity of money is not constant over time,
and policy recommendations based on such an assumption disregard an essential
aspect of economic and monetary processes. We shall return to this issue in Chapter 7.In the meantime,it will be useful to keep in mind that changes in velocity may

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

be due to any combination of the following factors:
(1) structure of payments flows, including financial flows,
(2) opportunity cost of money in relation to the yield of substitutes
appropriate for each segment of cash holders,
(3) technical factors, such as the efficiency of the payments mechanism and the degree to which payments are offset or obviated,
(4) use of deposit balances for the purpose of compensating banks
for services.
Velocities of money in the various sectors of the economy will reflect the specific
relevance of these various influences in determining the demand for money balance
in each sector, as well as cyclical and other differences in activity between sectors.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

3. The Flow of Payments
All sectors of the economy make payments in currency as well as by check. Only
purely financial transactions, including the flow of investment funds, and businessto-business transactions involve payment by check almost exclusively.

THE FLOW OF CONSUMER PAYMENTS

The level of money balances owned by households is determined largely by the
same factors that determine the level of corporate balances, i.e., the volume and
timing of payments and the relative attractiveness of alternative liquidity instruments.' Households (used here interchangeably with "consumers" and "individuals") generally try to keep bank charges at a minimum by avoiding unnecessary
activity in their accounts.
In paying many current household expenditures locally, the individual usually
has the choice of using either currency or check. Payments to distant points are
ordinarily made by check. Households that do not own checking accounts may use
postal money orders, bank money orders, or cashier's checks for long-distance
(and also local) payments.' Any increase in the use of such means for transferring
funds, like any greater use of currency, will tend to reduce the importance of personal expenditures in determining overall deposit velocity, but it will not be reflected to any significant extent in income velocity. It is not probable that changes
in the use of postal and bank money orders or similar payments instruments have

'Many of the devices adopted by business firms to improve their control over cash flows or to
spread out their work loads, such as cycle billing whereby department stores space out the billing of
customers over the month,automatically affect the pattern of consumer payments.
'In almost all Western European countries and in a number of other countries (about forty in all),
giro transfers provide the public with a cheap and efficient means of payment as an alternative to
checks. For a detailed discussion, see F. P. Thomson, Giro Credit-Transfer Systems (London, 1964).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

been large enough since the 1920's to affect demand deposit turnover ratios significantly, either in the long or in the short run. The rise in the volume of postal money
orders since 1929,for instance, has roughly paralleled the increase in debits.
In the normal course of events, individuals receive currency and checks at frequent intervals, mainly in payment of wages and salaries. Investment income and
Government and private transfer payments are also received by check at regular
intervals. Consumers can minimize the use they make of their bank accounts (and
of the resulting service charges) by cashing rather than depositing checks received. Many checks are endorsed more than once before being deposited. For
example, retail stores have traditionally cashed checks for their customers. Since
World War II, supermarkets have become a favorite check-cashing facility. Indeed, the volume of check cashing has assumed such proportions that, partly in an
effort to avoid delays at check-out counters, most supermarket chains have established special "service booths" or "courtesy booths" to cash the checks of their
customers (and, in some cases, to perform other services, such as the selling of
money orders and postage stamps as well as accepting utility bill payments). By
thus reducing the volume of recorded debits in relation to the actual volume of
payments made, the practice of endorsing over checks tends also to reduce the
importance of personal expenditures as a determinant of overall deposit turnover
rates.
For lower income groups especially, accounts at thrift institutions serve as a substitute for, or supplement to, checking accounts at commercial banks. Salary and
other checks received may be deposited in savings banks and savings and loan
associations (as well as in savings accounts maintained at commercial banks) and
withdrawals can be made as needed either in currency or in the form of cashier's
checks to make payments by mail. For obvious reasons, the thrift institutions and
savings departments of commercial banks normally discourage accounts of this
kind which frequently fail to build up large savings balances and serve only as
convenience accounts.' Nevertheless, some savings banks have in fairly recent
years sought to attract deposits by advertising "free checks", the number of which
is normally related to the size of the account.

'"Profile of Savings Deposit Rates", Business Conditions (Federal Reserve Bank of Chicago, July
1967), page 13.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

THE FLOW OF BUSINESS PAYMENTS

Most business transactions are paid by check. Nevertheless, business firms hold a
substantial portion of the currency that is in the hands of the public, but it is held
primarily for the purpose of paying wages and salaries (when they are dispensed in
currency) or as till cash (in retail and service establishments), and not for making
payments to other businesses or for paying taxes or dividends. Much of the
remaining currency held by business establishments represents the day's cash
receipts prior to their deposit in banks.
Practices and policies involved in the management of corporate cash balances
are as varied as the patterns of corporate cash flows. There are innumerable variants
in the basic pattern of matching the flow of payments with the flow of receipts and
for bridging temporal (seasonal and cyclical) gaps in these flows. The management
of corporate cash is one of the main responsibilities of the corporate treasurer. In
many large corporations, a special banking division (or a similarly designated unit
in the treasurer's function) is in charge of bank relations and cash management
policies. Meeting check flows requires supporting bank balances, and the adequacy
of collection and banking arrangements is analyzed systematically by large corporations and their bankers, with adjustments being made as needed.
It is convenient to consider business payments under two headings: those associated with current production and those representing intrafirm and other transfers
between accounts of the same economic entity.
A large part of all payments goes for the
purchase of intermediate products rather than to the agents of production. For
instance, since raw materials pass through several stages of fabrication before the
finished product reaches the final consumer, several transactions between individual business firms ordinarily take place; these transactions normally give rise to
check payments. Similarly, when finished products are distributed to final users,
several layers of distribution may be involved. As a result, the volume of check
payments related to the flow of goods and services alone exceeds by several times
the value of final output produced. Thus, in 1968, when total GNP amounted to
$860 billion, the volume of check payments may have reached $5 trillion. The proportion of payments directly related to current output is subject to cyclical as well
as to long-run fluctuations.
Any change in the composition of GNP may affect the volume of money payments required for final output. Consider, for example, the relative shift in demand
away from goods in favor of services. Goods pass through several stages of fabricaPAYMENTS RELATED TO PRODUCTION.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

The
tion and distribution, whereas services are usually sold directly to users.
growth of the service component in GNP (from 30 percent to 39 percent between
As a
1947 and 1967) tends to reduce the volume of intermediate transactions.
to
tends
result, the total volume of payments in relation to the value of final output
by
diminish, as does also the ratio of Vt/Vy. Final output purchased or produced
by
bought
products
finished
the Federal Government is another example. While
chargethe Federal Government are paid for with United States Treasury checks
from
able to accounts with the Federal Reserve Banks and therefore excluded
private
by
d
purchase
on
producti
reported debits, the payments for the factors of
producers who sell the finished products to the Government as well as the related
payments arising from the manufacture of intermediate products are, of course,
reflected in the debit totals. Furthermore, payment of taxes by individuals and
businesses may be regarded as charges against private accounts for goods and
an
services purchased by the Government for the public benefit. Nevertheless,
to
tend
will
output
s
economy'
the
of
increase in the Federal Government's share
that
is
reason
main
reduce debits in relation to the volume of national output. The
the Federal Government, in addition to purchasing a large volume of services,
usually buys directly from producers, so that the various stages of distribution are
normally eliminated.'
A large part of economic activity, in production and distribution, is carried out by corporations operating over extensive
geographic areas, often nationwide. Total corporate purchases alone are currently
But
equal to almost one fourth of estimated total debits at all commercial banks.'
of
activities
internal
by
a large portion of debits to corporate accounts is generated
corporations aimed at concentrating receipts from sales, transferring them to the
ily
points of disbursement, using funds in capital and credit markets, and temporar
represent
ly
presumab
investing redundant funds. As a result, total corporate debits
an even larger portion of total recorded payments than the share of the corpo-

INTRACORPORATE TRANSFERS OF FUNDS.

of GNP in 1965
4The rapid growth of the state and local government sector, absorbing 10 percent
in money turnover rates
as against only 5 percent in 1947, has also tended to retard the postwar rise
was 3.86 times
since this sector is one of low velocity. In 1956, corporate income velocity, for instance,
local turnover. R. T.
state and local velocity, and even consumer velocity was 1.35 times state and
Finance (Deof
Selden, "The Postwar Rise in the Velocity of Money: A Sectoral Analysis", Journal
cember 1961), page 539.
5As estimated in an annual release by the Board of Governors of the Federal Reserve System.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

rate sector in GNP. Management practices directed toward more efficient use of
corporate cash, including systematic investment of temporarily redundant cash,
tend to raise VY over time. These practices also augment the amount of intrafirm
transfers of funds in relation to the average cash balance and thus raise Vt as well.
The growing complexity of corporate structures—in part as a result of diversification of output and of mergers (in particular "conglomerate" mergers)—contributes
to multiplication of separate bank accounts. The tendency to decentralize operations geographically has the same effect. There are operating advantages in separating disbursement accounts (established mainly for accounting convenience and
to facilitate reporting and auditing) from collection accounts. Frequently, a corporation maintains several separate collection, disbursement, and special-purpose
accounts at the same bank. National corporations (which record the bulk of
corporate sales) usually have collection and/or disbursement accounts in most of
the localities in which they transact business. It is not unusual for a corporation to
maintain several hundreds of separate accounts, in perhaps as many banks. For
national retail organizations, the number of accounts may run into the thousands.
Usually the bulk of receipts is concentrated in a principal treasurer's account which
is the focal point of the cash management policies of the corporation. This account
may be split among several large banks located in New York City and, perhaps, in
some other principal centers, even though the head office of the concern may be
located elsewhere.
While the three-layer structure of collection, central treasury, and disbursement
accounts is fairly common in large corporations, there also may be additional layers
in each of the subsidiaries or divisional units. Instead of being concentrated at the
corporate headquarters, collection of remittances may be decentralized in a number of regional centers. Similarly, disbursement checks may be drawn against a
single treasurer's account, or the disbursement account may be decentralized to a
varying extent. Payrolls are typically charged against local disbursement accounts,
while checks in payment of raw materials or freight charges as well as for dividend
and Federal tax purposes are frequently drawn against central treasury accounts.
Some corporations transfer funds from collection accounts directly to local disbursement accounts in order to minimize cross-country transfers. Other companies,
to facilitate central control of cash flows or for other reasons, pass all collected
funds through their principal treasury account. Systematic (even daily) withdrawal of funds from collection accounts may be accomplished by using preprinted
depository transfer checks or by making proper arrangements through banks. The
centralization of corporate cash can be put on a semiautomatic basis by instructing

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

account or
the depository bank to transfer daily collections to a central or regional
ively,
Alternat
level.
d
to make such transfers when the account exceeds a specifie
local
ion's
an arrangement can be made whereby the manager of the corporat
simultabranch or sales office, when depositing cash and checks for collection,
te
corpora
main
the
of
one
holding
neously deposits a draft in favor of a bank
patterns
s
payment
and
accounts. In short, an infinite variety of corporate collection
patterns
exists. For our purposes, it is sufficient to note that changes in these
necessarily
affect cash balance needs, and thus Vy as well as Vt, but they do not
result in proportionate changes in these two series.
are made
The speed with which collections at, and transfers to, distant points
time element
influences the volume of the bank float and thus of net balances. The
s. Intrabalance
cash
for
in the collection process is therefore a factor in the demand
h
althoug
corporate transfers of funds between banks are usually made by wire,
. The wire
mails are still used for transfers to remote localities or for small amounts
e to all
availabl
been
transfer facilities of the Federal Reserve System have always
customtheir
member banks to transfer large amounts of funds for the account of
ence of
conveni
the
for
system
wire"
"bank
the
ers. The establishment in 1950 of
y
privatel
a
through
cities
ne
sixty-ni
leading banks now (1968) links 228 banks in
cash
te
corpora
of
operated network, and thus has further aided the centralization
facilities have
balances and their rapid transfer to points of disbursement. These
York or in
made money at distant points available almost instantaneously in New
a relawith
tions
any other money center of the country. Consequently, corpora
generally arrange
tively small number of collection (or "concentration") accounts
l accounts. On
principa
their
g
to have their balances wired daily to the bank handlin
day,the corporate
the basis of daily summaries prepared by these banks early in the
funds. The pooltreasurer decides on the required transfers or other disposition of
offset disburseto
units
ng
ing of funds enables surplus balances of some operati
facilitated
ments in others. In summary, this greater mobility of balances has
materially
application of scientific principles to cash management and contributed
intrafirm
each
Since
to the reduction of overall cash needs of national corporations.
another at
transfer of balances from one bank to another, or from one account to
volume of
the same bank, involves a debit (and credit) entry, thus swelling the
tion have
mobiliza
debits in relation to output and sales, techniques of money
tended to increase the volume of debits in relation to sales.
sector.
Interaccount transfers of funds are, of course, not limited to the business
govby
ned
s
maintai
account
Similar intraunit transfers take place among separate
onal
professi
ernment entities on the state and local levels. Wealthy individuals and
42

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

people may also maintain multiple bank accounts for convenience or business
purposes. The ultimate effect of pyramiding debits through interaccount transfers
in order to reduce average cash balances carried is to increase deposit velocity in
relation to income velocity.

THE FLOW OF SAVINGS AND INVESTMENT FUNDS

The savings and investment process generates large flows of payments made usually by check; the most significant exception is currency deposited by households
in savings accounts. Normally, the raising of funds in capital markets, trading on
stock exchanges or in over-the-counter securities, and transactions in existing real
or financial assets, including portfolio adjustments, involve payments by check.
In a growing economy with large savings, the volume of financial payments
generated in the process of floating and distributing new securities is large. The
underwriting of these securities by syndicates of investment bankers includes pooling funds and transferring them to the ultimate borrower. Debits totaling several
times the amount of financing involved may be recorded in this process and the
subsequent distribution of securities. Payments originating in financial institutions
are huge in relation to those involving physical outputs.'
The flow of savings into investment' may be direct (for instance, when business
firms reinvest undistributed profits) or through the purchase of corporate mortgages or other income-yielding financial assets. The financial assets thus acquired
(sometimes referred to as "primary" securities) represent direct claims of ultimate
lenders (such as individuals) upon ultimate borrowing units. The flow of savings
into investment may also involve the services of financial intermediaries who issue
to ultimate lenders claims upon themselves ("secondary" or "indirect" securities),
using the savings thus obtained to purchase "primary" securities issued by the
ultimate borrowers.
One of the main functions of financial intermediaries is that of concentrating
pools of savings and making them available to large users of funds. For organizations operating beyond a local scale, such activities often require intrafirm trans-

60ne leading financial firm recently advertised (The Wall Street Journal, June 26, 1968) that its total
payments in 1967 equaled one fifth of GNP.
71ncluding the acquisition of financial as well as real assets.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

fers of funds similar to those described above for nonfinancial corporations. Indeed, life insurance companies, the largest of the nonbank intermediaries, use a
network of local depositories and "concentration" accounts in much the same way
as do nonfinancial corporations. They,too, usually maintain separate disbursement
and investment accounts. A considerable volume of debits may also arise from
the process of pooling savings of individuals—as in the case of mutual investment
funds with their elaborate sales organizations. The widespread use of credit also
generates a large flow of charges to deposit accounts (and thus reported debits)
when credit is repaid.
Trading in existing claims reflects portfolio adjustments of institutional investors,
nonfinancial corporations, college and similarly endowed funds, and individuals,
including professional speculators. It takes place mainly on organized stock exchanges or over the counter. Various markets have different rules as to payment,
and some have arrangements for offsetting payments among participants. As a
result, changes in the share of transactions handled by the various markets and the
channels through which these transactions take place have a definite effect on
overall rates of deposit turnover as well as on the demand for cash of these
participants.
Among the various types of transactions in financial assets, stock market activities have tended to generate a sufficiently large volume of debits in periods of peak
activity (certainly so in the late twenties) so that they have an independent and
important influence upon rates of deposit turnover, at least in the main financial
centers. Indeed, it may be presumed, that throughout the 1920's trading on organized stock exchanges accounted for the bulk of financial transactions.
Indirect evidence suggests that since World War II stock market transactions
have been originating a considerably smaller proportion of total payments than in
the late twenties. Thus, in 1929, trading at the New York Stock Exchange was
equal to nearly 120 percent of GNP; in contrast, it was less than 10 percent of
GNP in each year prior to 1961, but it rose subsequently to 17 percent in 1968.
In general, during the entire postwar period this percentage has been much lower
than during the late twenties.'Similar comparisons hold for sales of corporate stock
on all national exchanges and over-the-counter markets. Between 1929 and 1961,
for example, stock trading on all exchanges and over-the-counter markets declined

8The New York Stock Exchange Fact Book, 1959 and 1968.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

in relation to GNP from 150 percent to 20 percent.' The value of listed corporate
shares outstanding is now only about one-half as large in relation to GNP as in 1929,
and the decline of the ratio of corporate bonds to GNP has been almost as large.
Individuals have increasingly relegated management of their savings to specialized institutions such as corporate pension funds, mutual funds, and trust departments of commercial banks. As a consequence, a much larger part of all corporate
securities is held by financial intermediaries (one third in 1961 against one fifth
in 1949 and only one fifteenth in 1929). On balance, the growing importance of
institutional investors (usually looking for long-term growth) in relation to professional and occasional traders (who, especially the former, may be at least
equally interested in short-term gains) is probably a principal factor in explaining
the relative decline in the turnover of outstanding securities, although since the
mid-1960's there has been a distinct increase in the rate of portfolio turnover by
mutual funds. The relative shift from stock to bond financing and the larger
amounts of private placements undoubtedly have had a similar effect.
Other types of financial transactions have not experienced the same decline
relative to the general level of economic activity (measured by GNP) as corporate
stocks. Compared with the late 1920's, for example, trading in Government securities (especially in Treasury bills) certainly increased when measured against
almost any index of business activity. It is only since World War II that the volume
of Federal Government securities has become very large by comparison with private
securities, and trading in United States Treasury securities has become the single
most important component of trading in financial markets. After the introduction
in 1929 of the Treasury bill as a money market instrument, and the inauguration
of the weekly issuance of ninety-day bills, active trading in outstanding issues began
to generate a large volume of debits. The introduction of longer term bills and the
widening use of coupon issues nearing maturity as a means of investing short-term
funds have contributed to creating a large volume of financial debits.
There is no way of measuring the total volume of payments associated with
trading in the various categories of securities or in specific financial markets and
their influence on velocity. Trading in stocks and bonds is reflected in the accounts
of dealers and brokers as well as of principals, such as individuals, insurance companies, and trust and pension funds,even though some part of the transactions orig-

9Report of Special Study of Securities Markets of the Securities and Exchange Commission, 88th
Congress, 1st Session, House Document No. 95, Part II, page 547.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

mating from portfolio adjustments gives rise to debits (and credits) to customers'
accounts with brokers rather than to the ultimate buyers' accounts at their banks.
Activities of dealers in Government securities, many of whom also make markets
of
or act as brokers for other money market instruments, generate a large volume
average
debits. Dealer accounts have exceptionally high turnover rates, because
bank balances maintained by such dealers are small in relation to the volume of
payments made. A special study conducted by the New York Clearing House Aswn
sociation shows that in February 1959 such accounts at the fourteen downto
New York banks turned over at an annual rate of 11,264 times (while accounts of
of
members of the stock exchange and of investment bankers turned over at a rate
which
in
way
peculiar
the
300 times a year)." This high turnover rate reflects
bank balances of Government securities dealers are used as clearing accounts.
Banks make and receive numerous deliveries of securities for the account of dealers, and receive or make the corresponding payments. Such payments will exceed
many times (about fifty in a working day, as the February 1959 figures show) the
by
amount of the opening balance in the account, but will normally be about offset
,
business
of
close
the
before
e
an equal amount of credits. "Day loans", repayabl
unbalance
also swell the volume of debits but will generally leave the closing
to
affected, unless the transactions fail to even out by an amount large enough
require an overnight loan.
The 1950's and 1960's witnessed the emergence of new types of financial claims
and the revitalization of markets in certain traditional instruments. It is quite likely
s
that in recent years financial debits reflect to a greater degree than ever payment
nts
instrume
market
arising from short-term investment of surplus funds in money
(not only by nonfinancial corporations but also by insurance companies, state and
Eurolocal governments, pension and trust funds, etc.) and foreign exchange and
less
much
a
is
which
activity,
dollar transactions rather than from stock market
significant contributor to financial debits than it once was.

dations Prepared for the
"New York Clearing House Association, An Analysis and Recommen
Requirements (Hearings before SubCommittee on Banking and Currency: Member Bank Reserve
House of Representatives,
committee No. 2 of the Committee on Banking and Currency, United States
261.
86th Congress, 1st Session, on H.R. 5237, Washington, D.C., 1959), page

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

4.The Measurement of Velocity
In principle, the computation of transactions velocity and income velocity is simple, but the practical problems encountered in measuring and interpreting both
magnitudes are many. This chapter is devoted to a discussion of the measurement
of these two time series. (See Chart 2).
As is frequently the case with derived economic time series, the underlying data
are collected to serve several purposes and hence do not measure perfectly the
magnitudes which the two definitions of velocity require. For instance, income
GNP
velocity
(V = .
is usually computed by relating total
7
Private money supply)
GNP, which includes Federal Government purchases paid with checks on United
States Treasury accounts at the Federal Reserve Banks, to the private money
supply, which excludes Treasury cash. On the other hand,the series on transactions
Debits
velocity
is derived from gross demand de(Vt =
Gross demand deposits;
posits which include a considerable amount of duplication. There are other inconsistencies, limitations, and problems of comparability, which affect long-run comparisons but which sometimes also raise questions with regard to the significance
of shorter run changes.
Exhaustive examination of the various series used to derive measures of Vy and
Vt cannot be undertaken within the framework of this monograph. A discussion
of the money supply series used in deriving V, is relegated to Appendix I, and
attention is there drawn to various statistical and analytical limitations of the currently used private money supply concept.
It will be observed that the Vt numerator encompasses a much greater range
of payments, including intrafirm movements of funds and transactions in real and
financial assets. On the other hand, the denominator is less comprehensive than
that of V,since it excludes the public's currency holdings. Both denominators exclude United States Treasury deposits. Alternative measures of Vy may be considered, and several such alternatives are discussed below.
V, has the statistical advantage of continuity and consistency over time, even
though the numerator as well as the denominator of the ratio include both estimates
and reported figures. Although the two underlying series are subject to frequent

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Chart 2. INCOME AND TRANSACTIONS VELOCITY, 1919-68
Times per
year

Times per
yeatRatio scale

40
/
///
..._ 3

Income velocity
Vy
Scale—.

it
1‘
•
30L
I
r%
I
•
%
.,#•'"••'
%
%
1,

,
/
,
I'
/
/"../
/
I

...,

20—

Transactions velocity\
Vt
%.
-.-- Scale

—2

/1-

V‘

• i/
N.
ge

11
1111
1111
1111
1111
1111
1111
1111
1111
1011111
1965 1968
1960
1955
1950
1945
1940
1935
1930
1925
1919
Sources: Based on data from the Board of Governors of the
Federal Reserve System; United States Department of Commerce;
and E. Oliver and A. J. Schwartz, Currelisy Held by the Public, the
Banks, and the Treasury.14Qfl1bly., December 1917- December 1944
Technical Paper No.4(National Bureau of Economic Research,1947).

by
revision, the continuity of coverage of the derived velocity ratios is not affected
which
such revisions. Not so in the case of V.The reporting sample of this series,
changes
several
e
undergon
has
data,
is based on reported rather than estimated
nts have
over the years in scope and coverage, and the required statistical adjustme
of overlap
been changed. The various time segments must be spliced on the basis
since
data (available in each case), and the resulting series for the entire period
conons
conclusi
broad
While
World War I is, in fact, not entirely homogeneous.
considerwith
it
from
cerning both long-run and cyclical fluctuations can be drawn
must be
able confidence as to validity, the statistical limitations of the V, series'
kept in mind.

in the Federal Reserve
'The various deposit turnover series mentioned on page 56 are published
Bulletin.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

INCOME VELOCITY

V,relates the current value of the aggregate final output of goods and services to
the private money supply (M), disregarding all intermediate payments required
to produce and distribute current output. Income velocity is not constant, either
over the business cycle or over the long run. Changes in the ratio of GNP
to M may reflect variations and lags in the relationships which link current output and the stock of money. Problems may also result from variations in the
structure of production and distribution, changes in the efficiency of the technology of payments, and fluctuations in the amount of cash balances required
to support transactions in existing real or financial assets (subsequently referred
to as "financial transactions"). Indeed, the claim that the relationship between the
aggregate flow of goods and services (measured by GNP) and the quantity of
money is fairly stable makes at least three implicit assumptions: (1) that the average number of intermediate transactions (as a product passes through various
stages of fabrication and distribution), each requiring payments between separate
units, remains constant over time; (2) that aggregate demand for cash balances
for all other reasons moves proportionately to that for meeting the flow of payments associated with current output; and (3) that the volume of transactions in
existing assets is also proportionate to the dollar value of current output. These
simplifying assumptions do not necessarily hold over time, as witnessed by the
time profile of V.and its relationship to V,(see Chart 2).
The private money supply excludes deposits of the Federal Government,
whether held at the Federal Reserve Banks or at commercial banks (with trivial
exceptions, such as accounts of disbursing agents in remote locations, where paying
by check redeemable at Federal Reserve Banks would involve considerable inconvenience). The justification usually given by various students for using the private
money supply to compute Vs—since there is no "official" series on income velocity
—is that the level of Federal Government spending is not related in any meaningful
way to the level of Treasury balances.
Two alternative measures of income velocity may be considered in an effort to
remedy the deficiency of having the contribution of the Federal Government appear
in the numerator of the ratio while the associated money supply is excluded from
the denominator. One is to limit the velocity ratio to the private product by removing from the numerator that part of GNP which is paid for by checks drawn on
United States Government accounts (Vi'). Alternatively, United States Treasury
deposits may be added to the denominator in order to establish the equivalence
between the numerator and denominator in the income velocity ratio (Vs
'
)
. Both

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

variants are shown in Chart 3.
The level of Treasury balances at commercial banks (except those rather stable
amounts maintained to compensate banks for services or in a disbursement account
at remote locations) reflects in the main the result of Treasury financing operations
and the collection of taxes (giving rise to deposits in Tax and Loan Accounts).
The argument in favor of including Treasury deposits at commercial banks thus
rests not on their relationship to the level of Government spending but on their
influence on the level of private deposits. These Treasury deposits occasionally
show what appear to be erratic fluctuations which are transmitted to VY, because
large amounts of private deposits are transferred to the Treasury account. While
these balances usually return within a relatively short period—via Treasury disbursements—to the private money supply, velocity values computed quarterly may
be affected by such shifts.
It is evident in Chart 3 that inclusion of Treasury deposits in the denominator
lowers the level of this variant of the velocity series (W),but cyclical variations
are quite similar to those in Vy. Indeed, very short-run shifts between private and
United States Treasury demand deposits do not affect quarterly totals to any great
extent, and the overall result of taking into account United States Treasury deposits
is to lower the velocity figures by about 4.4 percentage points.
Elimination of final United States Government purchases of goods and services
from the numerator has in most years the effect of lowering the value of the velocity
ratio V,,1 by comparison with Vy. In years of rapid expansion of defense expenditures, however, such as 1951-53 and 1965-68, the velocity measure \Ty.' rose less
rapidly than Vy. There is little justification for using Vy1. It is true that, when seen
from the product side of the account, such goods and services are paid by drawing
checks against accounts that are excluded from the money supply. But, when
viewed from the income side, all purchases by the Federal Government result in
additions to private balances of the sellers of these goods and services, and thus
the level of such balances is directly influenced by the amount of Government
purchases. Furthermore, while the finished products purchased by the Federal
Government are paid for in Treasury checks, private producers and suppliers
purchase the required factors of production by drawing on their (private) accounts.
Clearly in such periods a good deal of the stimulation of economic activity emanating from rising defense spending requires a considerable volume of intermediate
payments. Thus, the demand for private balances is raised even though the final
output is charged to Treasury balances. All intermediary sales of parts, semifinished products, and subassemblies which move through the various stages of
50

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

production also require coverage by adequate balances in private accounts and
result in charges to such accounts.
Other alternative measures of income velocity may be computed by widening
the concept of money to include time deposits at commercial banks, or even all
categories of time and savings deposits which their owners consider to be endowed
with the qualities attached to savings deposits at commercial banks. Finally, the
concept of money can be enlarged still further to include various liquidity instruments used by corporations, state and local governments, and other large transactors in order to economize on cash balances, as discussed in the preceding
chapter.
One such measure of income velocity (used, for instance, by Friedman and
Schwartz) widens the private money supply concept to include commercial bank

Chart 3. VARIANTS OF INCOME VELOCITY,1948-68
Times per
year
Ratio scale
4.54.0—

Vy4*

1948 49

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

50 .51

59 60 61 62 63 64 65 66 67 68
* Data plotted on an annual basis from 1948 through 1952,
thereafter plotted quarterly.
Sources: Based on data from the Board of Governors of the
Federal Reserve System and the United States Department
of Commerce.

time deposits(Vi').It shows a much smaller increase since World War II because
of the relatively more rapid growth of time deposits at commercial banks. In particular, following the "Accord" of 1951 commercial banks competed more actively
for savings and time deposits and in the sixties succeeded in improving their position relative to thrift institutions serving primarily households. Since the war, the
rate of growth of time deposits in commercial banks has substantially surpassed the
rise in demand deposits. At the end of 1966, time deposits of commercial banks
for the first time exceeded private demand deposits. Between 1947 and 1959, time
and savings deposits at commercial banks increased at an average annual rate of
5.7 percent, compared with a rate of growth in demand deposits of only 2.5 percent. Between 1960 and 1968,time deposits grew at an average annual rate of 22.6
percent while demand deposits rose at a rate of only 4.3 percent a year. As a result,
income velocity—computed as a ratio of GNP to money supply plus commercial
bank time deposits—in the latter period exhibited a much lower rate of increase
(whether United States Government deposits are included or not). This is true of
the period from the end of World War II as a whole and especially so during the
last decade when there has hardly been any increase at all.
Since the introduction of large-sized negotiable certificates of deposit (CD's)
in 1961, the new instrument has been obtained and held by many corporations and
other holders of large balances as a temporary investment of redundant funds,
and the amounts outstanding have grown rapidly. While some funds previously
invested in Treasury bills and other money market instruments were shifted into
CD's, the bulk of CD's issued represents, presumably, bank liabilities previously
shown as demand deposits. Furthermore, experience with CD's suggests that they
may be considered a closer substitute for demand deposits than other types of time
deposits. A variant of income velocity including CD's but no other types of time
liabilities(V)is also shown on Chart 3.
From an analytical point of view, the logical dividing line is clearly not between
demand deposits and time deposits at commercial banks, but rather between demand claims and time and savings deposits in any institution issuing such claims,
whatever its legal status and the terms used might be. Indeed, if a concept broader
than the private money supply is preferred, there does not seem to be any valid
reason to limit it to time deposits at commercial banks alone. Time deposits of
business firms are, no doubt, predominantly held with commercial banks, particularly since the banks began to issue marketable large-denomination CD's. This
instrument was also successful in drawing into commercial banks a very large part
of the time deposits of eleemosynary institutions and of municipalities that were
52

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

formerly held at savings institutions. But for individuals—and until the emergence
of CD's, for the units just cited—accounts at savings banks (in states where they
exist) and in savings and loan associations constitute an alternative to commercial
bank savings accounts, savings certificates, and similar instruments. The choice
made by savers depends mainly on rates available and convenience of location, and
only marginally on the legal status of the institution. We shall return to this issue
in the following chapter. An alternative measure of income velocity, which includes
time deposits at thrift institutions as well as at commercial banks in the denominator 07,1,is also shown in Chart 3. V,:z has been declining irregularly since 1951.

TRANSACTIONS VELOCITY

Transactions velocity should ideally measure the rapidity with which the economy's
total cash balances turn over. As a practical matter, it is computed as a ratio of
debits (in a large but limited number of centers) to average balances in demand
deposit accounts (other than interbank and United States Government accounts)
since no data on currency turnover are obtainable. Yet, currency has accounted
for about one fifth of the money supply in the post-World War II period as a
whole. Furthermore, the share of currency in the total money supply has been
subject to cyclical and long-run influences as well as to special influences arising
from conditions which existed during World War II and its aftermath.' There is no
reason to believe that efficiency in the use of currency has increased in proportion
to efficiency in the use of private demand deposits; indeed, the opposite might
be true.
Monthly transactions velocity data are derived as a ratio of two series, collected
for a large number of "centers": debits and demand deposits (Chart 4).
The amount of debits associated with the production and distribution of one
dollar's worth of final product is not constant over time. The volume of transactions in financial and real assets and of several other types of payments are only
loosely associated with the level of current output. Furthermore, some payments
are obviated through various kinds of compensating arrangements which in fact
may or may not affect significantly the demand for cash balances. Finally,
United States Treasury checks payable at Federal Reserve Banks, accounting for

2See Appendix II.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Chart 4. BANK DEBITS, DEMAND DEPOSITS, AND RATE OF TURNOVER
IN "OUTSIDE" CENTERS, 1919-68
Billions of
dollars

Billions of
dollars

Ratio scale

-1100

2000

.....*
11............

1000
800

WOII.
.....

Demand deposits
.......-Scale-.
•
I
I
I
/
Bank debits
/
/
-4-Scale
/
/

600

400

200

.........
.0, .......••••••

100rii-Tillitiii 1 I mm11111111111111111111111111111

Times per
yearl
30

Annual rate of turnover
20

11111111111
10 1111111111H1111111111111111111111111
1965 1968
1960
1955
1950
1945
1940
1935
1930
1925
1919
Source: Board of Governors of the Federal Reserve System.

the bulk of Federal Government payments, are not included in reported debits.
Current data on V,have been published monthly for almost half a century in the
Federal Reserve Bulletin (as well as in separate releases) together with debits from
which they are derived. The coverage of debits has been gradually extended to
include entire metropolitan areas, and estimated debits (less than one tenth of the
total) were added for banks not actually in the reporting sample.' As a result of
the most recent revisions,' the most inclusive series now reflects deposit activity in

3For a detailed discussion of the various published and historical velocity and related series, see
George Garvy, Debits and Clearing Statistics and Their Use (Board of Governors of the Federal
Reserve System, revised edition, 1959), Chapter VII.
4Undertaken in 1964 and 1966. See Federal Reserve Bulletin (May 1965 and March 1967).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

233 standard metropolitan statistical areas;it is estimated to cover more than three
fourths of all relevant debits in the country.
In terms of numbers and amounts, the overwhelming bulk of debits represents
payments by check or similar forms of settlement (for example, by draft). Large
payments at distant points frequently involve wire transfers. Reported debits make
no distinctions with regard to the underlying transactions as long as a charge to the
depositor's account arises. All transactions in real or financial assets, payments
arising from the extension of credit,and the raising of funds in capital markets, as
well as all transfers between separate accounts of individual economic units and
the exchange of one kind of money (demand deposits) into another (currency)
or into savings deposits (at the same institutions or elsewhere) or other near
moneys, give rise to debits, as do also the various charges for bank services and
purchases of securities through the banks. Nor are any other distinctions made
between various types of deposit accounts or holder categories: special checking
accounts, for instance, are pooled with regular personal and business accounts.
Published rates of turnover (seasonally adjusted) are, in fact, average rates at
which the various components of private demand deposits are used; a portion of
the total is turned over very intensively, while another portion, such as compensating balances maintained by business firms, is virtually at rest. Liquidity reserves
and temporarily redundant funds of business firms and individuals, as well as of
municipal governments and eleemosynary institutions, also have a relatively low
velocity.
To simplify computations and to speed up the release of statistical data, rates
of turnover are computed from gross deposits as they appear in bank records
instead of from net deposits (gross deposits minus bank float). Also, bank float
cannot be allocated to individual banks and, therefore, it is not possible to estimate
net deposits of the banks in any standard metropolitan area or city in computing
turnover rates for such areas. Actual rates of net deposit turnover are thus significantly higher than the published rates. It is likely, however, that the ratio of gross
to net deposits has been sufficiently stable to justify the use of the available velocity
series for at least short-run comparisons, even though at certain times transportation and/or processing delays may have resulted in fluctuations in the ratio of float
to gross demand deposits. Some of the changes in the technology of collecting
checks that are likely to have affected the ratio of net to gross deposits in the long
run are discussed in the following chapter.
Detailed analysis of Vt for individual areas suggests that, in financial centers,
payments related to securities and similar assets transactions tend to raise turn-

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

over rates, since accounts of many categories of financial firms have very high turnover rates (Chart 5).In particular, turnover rates of demand deposits in New York
City reflect to an unusual degree purely financial activities, including transactions
arising from dealings on the New York Stock Exchange as well as in other securities
markets.' This is true, though to a less extent, for other leading financial centers.
As a result, since 1953 data covering the six leading financial centers' have been
shown separately, in addition to those for New York City (which are available as
far back as 1919). For certain purposes, the behavior of velocity in the separate
group of financial centers is of interest. Our subsequent discussion relates, however,
unless otherwise stated, only to the national series on velocity rates in the 226
"other" areas, excluding New York City and the six other financial centers. The
series for the 226 centers also contains a large volume of purely financial transactions,' since some of these centers are in many respects similar to those in the
six-center series.
In view of the considerable differences in turnover rates of individual accounts,
plans to relate reserve requirements to deposit velocity have been advanced since
the early thirties' when such a proposal was developed as the result of a study
directed by Winfield Riefler. The Riefler proposal involved basing reserve requirements for individual banks on total debits to total deposit accounts, subject to a

'For a discussion of the particular factors affecting velocity in New York City, see George Garvy,
Op. cit., Chapters III and VII.
'Boston, Philadelphia, Chicago, Cleveland, Dallas, and San Francisco. While these centers may be
regarded as the principal financial centers outside New York City for the post-World War I period
as a whole, other centers are now of similar if not greater importance. To a certain degree, the choice
made in 1953, when this series was initiated, was arbitrary.
7For an alternative attempt to obtain a measurement of the flow of checkbook money payments
of identical
that would be as free as possible from debits originating in shifts between separate accounts
holders and from the transformation of demand into time deposits and financial assets of transactions,
Analytical
see the unpublished Ph.D. dissertation by Anthony Mach, The Institutional, Statistical and1967).
The
Importance of Federal Reserve Check Collections (Boston College, Graduate School,
in
author proposes that check collections through Federal Reserve channels be "used as a numerator
velocity studies of the transactions demand for money"(page 5). Mach found,for the period 1952-62,
a stable, almost one-to-one relationship between the volume of checks collected through the Federal
Reserve System (excluding Treasury checks) and "nonfinancial transactions" as estimated by
McGouldrick (see page 77, f. 1). The ratio of Federal Reserve check collection to total debits fluctu1944 and then
ated in the period of 1920-64, rising from 25.6 percent in 1920 to 58.6 percent inpurely
financial
declining to 29.1 percent in 1964, with a strong suggestion that the changing share of
debits significantly influenced this relationship.
'See Warren Smith, "Reserve Requirements in the American Monetary System", in Commission on
Money and Credit, Monetary Management(Englewood Cliffs, New Jersey, 1963) and Neil H. Jacoby,
"Bank Legal Reserve Requirements", Banking and Monetary Studies, ed. Deane Carson (Homewood,
Illinois, 1963).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Chart 5. RATES OF TURNOVER OF DEMAND DEPOSITS,1943-68
Times per
year
140 —

120

100

60
New York City
40
Six financial centers
20
226 other centers

1944

1
1946

1
1
1948

1
1950

1952

1
1954

I
1
1956

1
1
1958

1960

1
1
1962

1
1
1964

1
1966

1
1968

Data plotted monthly.
Source: Board of Governors of the Federal Reserve System.

maximum limit. A later proposal, made by Jacoby and Norton' more than a quarter of a century later, would set reserve requirements for a limited number of
groups of banks (say, three or five) classified by average debits to total deposit
accounts from which identifiable classes of deposit accounts involving mainly
financial transactions would be eliminated. Its rationale for basing reserve requirements on deposit velocity was to use the best possible approximation of differences
in the income velocity of deposits. The Federal Reserve Board would be expected
to set and vary the statutory reserve ratios (not necessarily for all categories simultaneously) in the light of policy objectives.

°Frank E. Norton and Neil H. Jacoby, Bank Deposits and Legal Reserve Requirements (Los
Angeles, California: UCLA Graduate School of Business, 1959).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SECTOR VELOCITIES

Turnover velocity of deposits may rise because of a change in the volume of payments in relation to the volume of deposits held, or for purely technical causes.
It may also fluctuate because of a change in the relative weight of various holder
categories or economic sectors with typically different velocities. As one might
expect, velocity differs widely among sectors. Velocity in the consumer sector, for
example, is lower than in the corporate sector, and cash turns over faster at larger
than at smaller corporations. Velocity fluctuations over time also show considerable differences for individual sectors.
Indeed, any aggregate measure of velocity—whether of income or transactions
—is a composite of millions of money turnover ratios for individual consumers,
businesses, and other holders of cash balances. Total velocity may thus be viewed
as the weighted average of velocities of single accounts or economic units. When
individual accounts are grouped according to some analytical principle, such as
industry branches or economic regions, velocity of the total money supply may be
viewed as a weighted average of a number of differing sector velocities."
Structural factors may help to explain the cyclical as well as long-run behavior
of velocity. Thus, changes in the distribution of cash balances between sectors
and/or changes in the relative share of individual sectors with different velocities in
total money payments can exert an influence on aggregate velocity. Part of the rise
in total income (or transactions) velocity during periods of cyclical expansion
might be explained by a greater relative gain in the volume of transactions in highvelocity sectors. Conversely, the decline in velocity, or a slower growth of velocity,
during recessions might be explained in part by sharper falloffs in income (or the
volume of transactions) originating in the same high-velocity sectors. Cyclical
variations in aggregate turnover rates might also be explained in part by relative
shifts in money holdings between sectors over the course of the cycle. The findings
of one study suggest, indeed, that shifts in the relative weight of the corporate and
consumer sectors are significant in explaining cyclical velocity swings."

which quarterly esti"Studies of sector velocity are usually based on flow-of-funds estimates, from
out) are available since
mates of money balance and spending (with various degrees of netting
data for cash and sales or
World War II for the principal segments of the economy. CorporateSee,
for instance, Richard T.
revenue have also been used for measuring sector transactions velocity.
Journal of Finance
Selden, "The Postwar Rise in the Velocity of Money: A Sectoral Analysis",
Federal Reserve
(December 1961) and Paul F. McGouldrick, "A Sectoral Analysis of Velocity",
Bulletin (December 1962).
"See Paul F. McGouldrick, op. cit., pages 563-65.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

S. The Statistical Record
CYCLICAL FLUCTUATIONS

Fluctuations in velocity over the economic cycle reflect the more intensi
ve and
efficient use of money during periods of rising activity as a result
in part of the
higher opportunity cost of holding money. The close relationship betwee
n the level
of business activity and velocity was first recognized in the early
1920's. Turnover
rates of total deposits from 1870 to 1914, estimated by Carl Snyder
from related
data on bank clearings, clearly indicated a pattern of cyclical fluctua
tions during
that period. For a time,an "index of the turnover of bank deposits"
computed at the
Federal Reserve Bank of New York was widely used as an indicat
or of business
activity.' After World War II, the rate of deposit velocity continued to show
a definite cyclical pattern. Measures of income velocity, available for a shorter
period
and on a quarterly basis only, also reflected cyclical expansions and
contractions
(see Chart 6).
Tables 1 and 2 demonstrate that income as well as transactions
velocity has
risen substantially in each period of expansion since the end of World
War II, and
has either declined very slightly (usually by 1 percent to 3 percent, and
never by
more than 5 percent) or remained essentially unchanged in each contrac
tion. The
moderate nature of velocity declines and the unchanged velocity
in some periods
of recession are explained, in part, by the upward trend in velocity that has characterized the whole post-World War II period and, in part, by the relative mildne
ss
of postwar recessions. Since the money supply has expanded almost continuously
from the end of World War II, cyclical fluctuations in the two velocity indexes (and
in the variants shown in Table 1) reflect, in the main, more rapid increases in GNP

'Carl Snyder, "Deposits Activity as a Measure of Business Activity"
, Review of Economics and
Statistics (October 1924). See also J. W. Angell, The Behavior of Money(New
York, 1936), especially
Chapter IV.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

LD WAR II
Chart 6. INCOME VELOCITY OF MONEY IN THREE POST-WOR
BUSINESS CYCLES
Times per
year
4.64.44.21960-68

4.03.8-

^
3.6-

e•
'
^
•

1957-60

-1

3.2 —
1953-57
3.0 —
2.8
I

1
—5

lit
5

Quarters from trough
I
I I I I I
15
10

I I
20

l(f lit
30
25

Zero line indicates troughs in business activity, as determined by
the National Bureau of Economic Research: 3rd quarter 1954, 2nd
quarter 1958, 1st quarter 1961.
Sources: Board of Governors of the Federal Reserve System
and the United States Department of Commerce.

checkbook money
and in the volume of debits than in private money supply or in
s contractions
in periods when the economy was expanding. In the four busines
fluctuated
debits
while
ged,
since 1946, GNP declined or remained virtually unchan
as well as
supply
within a very narrow range. The private (and the total) money
s in no case
demand deposits also tended to remain fairly stable, with change
tion.
exceeding 3 percent during any period of contrac
intensively and
When business activity expands, the money supply is used more
sure, businesses,
supports a proportionately greater volume of transactions. To be
s, partly by
holding
cash
their
to
individuals, and governmental units tend to add
of money
cost
the
when
borrowing additional funds from banks. At the same time,
s, to find better ways
is higher, they redouble their efforts to economize on balance
60

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

to coordinate their receipts with payments, to expedite collections, and to reduce
idle funds. Since capital formation is more volatile over the cycle than other spending flows, rising activity is accompanied by a more than proportionate increase in
capital expenditures and in the various activities related to their financing. Availability of new securities tends to give rise to chain reactions in portfolio adjustments, as do rising activity and price expectations associated in the equities markets. The influence of financial transactions shows up most clearly in measures of
deposit velocity where, relative to GNP, the more pronounced upward thrust in
debits during periods of cyclical expansion (Table 2) is due in part to increased
financial activity, even in the series from which the main financial centers are
eliminated.

TABLE 1

CYCLICAL CHANGES IN GROSS NATIONAL PRODUCT,
MONEY SUPPLY,AND INCOME VELOCITY,1946-69
Percentage changes between troughs and peaks

Cyclical phase

GNP Money
GNP adjusted* supplyt

Income
Money velocity
supplyt
(Y1)

(2)

(3)

(4)

40
4
32
3
24
3
17
1
90

+ 6
— 1
+16
+ 2
+ 5
0
+ 2
+ 1
+41

+ 6
0
+15
+ 3
+ 4
+ 1
+ 2
0
+40

+
—
+
—
+
—
+
—
+

+388

+89

+90

+157

(1)
1946 I —1948 IV (expansion)§
1948 IV-1949 IV (contraction)
1949 IV-1953 II (expansion)
1953 II —1954 III (contraction)
1954 111-1957 III (expansion)
1957 111-1958 II (contraction)
1958 II —1960 II (expansion)
1960 II —1961 I (contraction)
1961 I —1969 IV (expansion)

+
—
+
—
+
—
+

34
3
44
1
22
2
15
0
+ 89

+
—
+
+
+
—
+
—
+

Entire period, 1946 1-1969 IV

+385

Income
velocity
(Y2)

(6)

(5)
26
3
24
2
17
2
13
1
34

+
—
+
+
+
—
+
—
+

32
3
14
1
19
3
15
2
35

+158

Note: Cyclical phases shown between peaks and troughs, as determined by the National
Bureau of
Economic Research, are based on seasonally adjusted series except for column (4). Because
of lack
of monthly GNP figures, analysis on a monthly basis is not feasible.
*GNP minus Federal Government purchases of goods and services.
tDaily average series of private demand deposits plus currency.
tColumn (3) plus United States Government demand deposits (not seasonally adjusted).
§Instead of the 1945-IV trough for which GNP data are not available, the expansion
is measured
from the first quarter of 1946. Money supply.for 1946-1 is estimated
from the end-of-the-month rather
than from the daily averages series.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

TABLE 2

CYCLICAL CHANGES IN DEMAND DEPOSITS,DEBITS,
AND TRANSACTIONS IN VELOCITY,1945-69*
Percentage changes, seasonally adjusted series

Cyclical phaset
October 1945—November
November 1948—October
October 1949—July
1953—August
July
1954—July
August
1957—April
July
1958—May
April
1960—February
May
February 1961—December

Debits to
demand
deposits
+ 63
1948 (expansion)
— 4
1949 (contraction)._
+ 50
1953 (expansion)
+ 1
1954 (contraction)
+ 29
1957 (expansion)
— 2
1958 (contraction)
+ 20
1960 (expansion)
+ 1
1961 (contraction)
+151
1969 (expansion)

Entire period, October 1945—December 1969

+821

Demand
deposits
adjusted:
+ 21
0
+ 24
+ 1
+ 8
0
+ 5
+ 2
+ 50
+165

Transactions
velocity
+ 35
— 5
+ 21
+ 1
+ 20
— 2
+ 14
0
+ 67
+247

from 1964
*Through 1963, in 337 centers apart from New York City and six other financial centers;
in 1964.
on, 226 centers. The two series have been spliced, using the ratio of 226/337 centers
Research.
tBetween peaks and troughs, as determined by the National Bureau of Economic
of collection.
Wther than interbank and United States Government deposits, less cash items in process

Many of the economies in the use of money, adopted in a period of limited
availability and rising costs, are not reversed in periods of easy money. The experience gained in investing excess cash in money market instruments or incomeyielding savings accounts in periods when interest rates are high is not abandoned
in times when the attractiveness of such investments is lessened; nor, for example,
are the techniques developed to reduce mail and collection floats. Indeed, advances
in the "technology" of managing money,stimulated by rate and availability factors
during expansions, have a lasting effect and become a long-run influence affecting
velocity.

LONG-RUN CHANGES

The admittedly deficient income and transactions velocity data for earlier periods suggest a declining trend during the nineteenth century and first decade of this
62

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

century.' Numerous series have been constructed to study long-run trends,' even
though for the years prior to 1919 such series have necessarily been based on somewhat doubtful assumptions and interpretations. Many of the contradictions in the
conclusions reached by different investigators as to the existence and direction of
a secular trend in velocity result from disagreements in defining money supply,
estimating income and transactions, and time periods covered. Of course, the more
the analysis is extended into the past, the more uncertain the estimates of the relevant magnitudes become.
In particular, the further we move into the past, the more the dividing line
between demand and time deposits becomes blurred. Prior to the banking reform
of 1933, which among other changes prohibited payment of interest on demand
deposits, the difference between time and demand deposits was less sharp than in
subsequent years.' For instance, before the establishment of the Federal Reserve
System in 1913, legal reserve requirements did not distinguish between the two
types of deposits. Banks and supervisory authorities, therefore, had no strong
reasons for enforcing proper classification and reporting; indeed, banking statistics
for these years include a large volume of unclassified deposits. Moreover, many
students believe that in subsequent years, perhaps until the banking reform of 1933,
a substantial volume of demand deposits was incorrectly classified as time deposits.' Also, the relationship between clearings (alone available prior to 1919)
and debits becomes more and more uncertain the further one goes back,since origi-

2See, for example, Clark Warburton, "The Secular Trend in Monetary Velocity", Quarterly
Journal
of Economics (February 1949), and Milton Friedman and Anna J. Schwartz, A Monetary
History of
the United States, 1867-1960(Princeton, New Jersey, 1963).
3A number of these studies are conveniently tabulated and summarized by Richard T. Selden,
"Monetary Velocity in the United States", Studies in Quantity Theory of Money,
ed. Milton Friedman (Chicago, 1956).
4Consider,for example, the following dialogue taken from the Hearing held before the United
States
Senate's Banking and Currency Committee in 1913 (Volume III, pages 2370-71),
on the bill to
establish the Federal Reserve System:
Senator Nelson (Minnesota):"Do you issue time certificates of deposit?"
Mr. Ingle (V
resident of a large Baltimore bank): "We issue demand certificates of deposit
bearing interest.....
Senator Nelson: "Is not that a specie of savings?"
Mr. Ingle: "No, sir; not in our connection.... It is only in the case of a man who conies in
say, $10,000 or $20,000, for which he has no particular use for sixty or ninety days or with,
months ... and feels that he ought not to let it lie in a bank earning nothing; we will then four
give
him a demand deposit at, say, 2 or 2/
1
2 percent. I have known us to pay as much as 3 percent,
but not habitually."
5See, for instance, Warburton, op. cit., pages 81-84.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

ions
nally clearings were reported only by large centers in which financial transact
estiGNP
y,
Similarl
dominated and additional cities were added only gradually.'
based
mates required to compute income velocity for the nineteenth century are
on some heroic assumptions.
After a
The course of velocity movements since 1919 is shown in Chart 2.
decline during the postwar recession year of 1921, both turnover rates(V,and V,)
1920's
increased fairly steadily through 1929. Income velocity rose during the
level
1929
the
that
result
considerably slower than transactions velocity, with the
d
exceede
V,
of V.just about equaled those experienced in 1919 and 1920, while
the two
its post-World War I level by a wide margin. The differential behavior of
The
plotted.
is
V,
to
V,
of
ratio
velocity series is illustrated in Chart 7, where the
in
surge
the
measure
greater rise in deposit velocity in the 1920's reflects in some
financial market activities.
decreased
During the initial years of the Great Depression, both velocity rates
d in subcontinue
sharply while business activity rapidly contracted. The declines
Dursequent years (through 1945), as the economy remained excessively liquid.
than
faster
grew
supply
money
ing these years also, the volume of deposits and
on
producti
business activity, whether measured by GNP or by debits. Even though
II, the
and all other aspects of business activity increased greatly during World War
d to
continue
g
financin
war
of
huge volume of deposits created in the process
lows
all-time
to
depress rates of income velocity and deposit turnover, which fell
in 1946 and 1945, respectively. V,declined, however,less sharply than Vy, and the
ratio shown on Chart 7 rose during World War II.
of
As business activity continued to expand during the postwar period, the rate
rapidly
growth of the money supply first slowed, then speeded up somewhat more
between 1953 and 1961, and finally accelerated in the period 1961-68. While
II
both income and deposit turnover rates increased in the post-World War
of
level
The
decades, the latter advanced more quickly (Chart 5) after 1953.
8 times
V, was 6.5 times that of V, in 1953, and by 1968 it was more than
had
ent
managem
money
of
art
the
6,
Chapter
greater. As discussed more fully in
monein
y
flexibilit
Greater
undergone a technological revolution since the 1950's.
g the
tary policy and the related increased activity in the money market followin
"Accord" of 1951 tended to increase the volume of financial payments in relation

6For details, see the monograph quoted on page 54,footnote 3.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

to GNP.Rising interest rates on debits turnover and technological advances in making and collecting check payments tended to reduce the measure of money used in
the denominator of both velocity ratios, but transactions velocity was further increased by additional debits generated by the investment of redundant funds in
financial markets and intrafirm flows designed to increase efficiency in the use of
corporate funds. For the entire period since the end of World War II, income
velocity increased from 1.97 in 1946 to 4.59 in 1968, while transactions turnover
rose in the same period from a low point of 12.8 in 1945 to just over 36.5 in 1968.
The movements of the additional variants of Vy, shown in Chart 3, indicate quarterly changes that are not very different from those of V,in the main series.
The rising trend of aggregate velocity characterizing the postwar decades has
been broadly diffused throughout the economy. Whether one looks at the consumer
sector or the business sector, manufacturing corporations or public utilities, small
firms or large firms, companies producing automobiles or companies engaged in
trade, the trend in income velocity during the years since the end of World War II

Chart 7.
W19-68

RATIO OF TRANSACTIONS VELOCITY TO INCOME VELOCITY

Vt
Vy

5 1 1 1 1 1
1919

1 1 1 1

1925

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

1111

1930

11_11
111111111
1111
1111
1111
11
1940
1935
1945
1950
1955
1960
1965 1968

has been distinctly upward. This postwar rise does not have its roots primarily in
the structural changes taking place among the different sectors of the economy.
Rather, the rates of money turnover have increased for nearly all components,
indicating a pervasive decline in cash balances relative to income and payments
flows. Indeed, longer run structural changes (in distribution of cash holdings and
in share of transactions) have worked, on balance, in the opposite direction—i.e.,
toward retarding the rise in velocity.'
The almost continuous and rapid rise in both types of measures of velocity since
the end of World War II, which by the end of 1968 carried V,to 233 percent of its
level at the post-armistice trough while Vt rose 285 percent, is in striking contrast to
the record of the earlier period.It is also not consistent, as we shall see in Chapter 7,
with some interpretations of the forces underlying the demand for cash balances
which have been advanced in recent years.

7See Paul F. McGouldrick,"A Sectoral Analysis of Velocity", Federal Reserve Bulletin (December
1962).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

6. Factors Affecting Velocity
Factors affecting rates of money turnover in the long run can be grouped under
four headings: first, compensating arrangements to settle payments which tend to
reduce the demand for cash balances as well as the volume of debits; second, important and rapidly developing policies and arrangements designed to decrease the
amount of balances required to meet a given flow of business payments; third, the
related efforts of corporations to invest temporarily redundant cash; and, finally,
recent developments that tend to lessen the demand for cash balances on the part
of consumers.

COMPENSATING ARRANGEMENTS

Arrangements which aim at obviating the need for making payments tend to reduce
the volume of check payments and the need for supporting balances. Most arrangements of this nature involve the mutual offsetting of credits and debits, usually
through the intermediary of a clearing agency. Some of these clearing agreements
date back to the nineteenth century; others are of relatively recent inception.
The oldest and most rudimentary form of compensation—one which does not
involve any special arrangements—is used mostly in rural areas by merchants who
extend credit to farmers and in turn credit their accounts as produce is purchased
from them. A certain amount of netting out takes place in many interbusiness
transactions. The largest volume of offsetting arises in the activities of organizations
that are created specifically to clear payments stemming from certain activities
which generate a large volume of transactions among a relatively small number of
participants. For example, only a small proportion of the payments among brokers
that result from trading securities on organized stock exchanges and/or dealings on
commodity markets requires drawing checks. If these offsetting arrangements did
not exist, the volume of financial payments,including a large volume of transactions
in commodity futures, would constitute a much larger share of total debits than
they currently do.
Arrangements to offset payments arising from trading on the New York Stock
Exchange were initiated as early as 1880. Similar clearing arrangements have also
been developed and perfected over the years in other exchanges and industries

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

where frequent transactions between a limited number of firms give rise to numerous payments or credit-and-debit transactions. This process has by no means
been terminated. In 1969, clearing of securities traded in the over-the-counter
market was inaugurated. Trading in United States Government securities takes
place over the counter rather than on organized exchanges. There is still little
netting out of payments in the Government securities market and in the over-thecounter market for corporate securities.
Clearing arrangements may involve not only financial transactions but also payments among firms within certain industries. Important examples of clearing
arrangements may be found in various segments of the transportation industry,
such as railroads, trucking, and airlines. These compensating arrangements may be
operated through banks acting as agents or on their own.
While the specific details differ, the basic principle of compensating arrangements is always the same: part of the gross payment procedure is obviated by an
offsetting process among the participants. The result is that the net monetary
amount of the checks written in settlement of balances is only a part, usually a
small part, of the amount of the gross transactions and thus requirements for
transactions balances are reduced. Consequently, offsetting arrangements tend to
increase V. Whether on balance they increase or reduce V,, depends in each case
on the alternative payments arrangements which they replace.
In contrast to transactions among members of organized exchanges and closely
related business firms, netting out in payments by individuals is negligible. Important exceptions are customer balances carried by stockbrokers in those cases
where securities purchases and sales and borrowing for margin purposes and related
repayments (and in some instances receipts of dividends) are offset on the books
of brokers so that the only resulting net differences are debit entries to customer
(or broker) checking accounts.

ECONOMIZING ON CORPORATE CASH

The rise in the cost of money since the end of World War II has put pressure on
corporate treasurers to reduce their cash requirements to a minimum. At the same
time, demand for credit and the corresponding vigorous steps taken by commercial
banks to attract funds through the introduction of additional financial instruments
have offered new possibilities of earning attractive returns to holders of excess
cash. The endeavors of corporate money managers to keep cash holdings shaved
as closely as possible to anticipated needs, and the widespread application of scien68

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

tific management principles to the utilization of cash, have tended to increase V.
At the same time, economizing on cash balances also tends to increase the volume
of financial transactions and of related debits. Since temporarily redundant funds
are usually invested for very short periods, a systematic corporate policy to restrict
bank balances to a minimum generates a considerable volume of financial debits as
redundant funds are invested and reinvested in money market instruments, thus
increasing turnover rates of corporate cash balances by increasing the volume of
debits while balances are reduced.
The size of aggregate corporate cash balances is essentially determined by the
volume of corporate payments. However, this relationship varies from industry to
industry and from firm to firm, as well as over periods of time. Obviously, conditions differ widely among industries and individual companies, depending on
seasonal patterns of production, purchasing of materials and sales of final products,
billing and payments patterns, the degrees of vertical and horizontal integration,
reliance on credit to meet seasonal needs, and other factors, some of which are
peculiar to an industry (or even to a company) while others have wider applicability.' Also, liquidity needs of corporations differ extensively according to the
structure of markets in which they operate as buyers and sellers, their corporate
size, respective financial policies, capital structure,long-term indebtedness, current
and future plant and equipment expenditures, and other factors. The corporate
velocity ratio is also influenced by the need to compensate depository banks for
their services and to maintain balances adequate for establishing banking relations
which can assure access to credit when needed.
Modern cash management techniques originated in the 1920's, when they were
but one phase of a trend toward scientific business management, by seeking new
ways to employ temporarily redundant funds. Indeed, corporate funds were substantially used in financing the stock market boom of the late twenties.' Low interest rates in the 1930's and during World War II, however, offered limited incentives
to justify maintenance of staffs required to keep redundant corporate funds continuously invested in money market instruments.

'See Morris A. Copeland, A Study of Moneyflows in the United States(New York:
National Bureau
of Economic Research, 1952), Chapter II, and Richard T. Selden, "The Postwar
Rise in the Velocity
of Money:A Sectoral Analysis", Journal of Finance(December 1961).
'Loans to brokers by the weekly reporting New York City member
"for others"—mostly
corporations—rose from $579 million at the end of June 1926 to a peak banks
of nearly $4 billion during
the first part of October 1929.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

The situation changed rapidly in the postwar years, particularly after the unpegging of rates on Government securities. At the same time, the growing preference of banks to make explicit charges, instead of requiring compensating balances, to business (and other) depositors for the continuously widening range of
services they offered and the general trend of corporate management to apply more
scientific methods to all (including financial) aspects of corporate business stimulated new efforts by corporations toward a more efficient use of cash. This development was encouraged by other factors, such as the increase in corporate tax rates
(with corporations regularly accumulating funds in anticipation of tax payments)
and the availability of a wider range of money market instruments.
Pioneering efforts to economize on cash balances have spread rapidly through
the business universe. The relatively small number of large corporations handling
the bulk of business in manufacturing, public utilities, transportation, and communications, and increasingly in retail trade, explains why ways of handling cash
developed by one became generalized fairly rapidly among other leading corporations. Efforts of commercial banks in financial centers to attract corporate accounts
by assisting corporate treasurers toward efficient cash management (as well as
various complementary endeavors like the seminars of the American Management
Association on financial management) have given added impetus in recent years
to the trend toward economies in corporate cash balances. Banking concentration,
especially among city banks, which eliminates the need for corporations to maintain two or more balances in one area for goodwill reasons, accomplishes the same
effect.
The trend toward greater efficiency in the use of corporate balances is portrayed
in Chart 8. No series on total corporate payments is presently extant. The two
principal indirect measurements available are "nonfinancial transactions" and
"gross sales". The former involves complex and somewhat arbitrary estimating
procedures, but it is a more inclusive measure covering all payments except those
of a purely financial nature and those for existing real assets. The sales-to-money
ratio is a measure of nonfinancial corporate velocity derived from accounting
records. As financial transactions have undoubtedly increased more rapidly than
real transactions, mainly because of the payments that are related to the investment
of excess cash, total corporate payments have increased even more in relation to
money holdings than is shown in either of the series.
While there is abundant evidence of corporations economizing on money holdings, it is also clear that overall liquidity ratios have been relatively more stable.
Interest-bearing claims have been increasingly substituted for demand balances in
70

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

corporate liquidity portfolios (more fully discussed in Chapter 2). The extent to
which various nonmonetary liquid claims (e.g., CD's and short-term Government
securities) have been substituted for cash is evident in corporate financial statements which frequently group "cash and invested cash" under a single item.
Efforts to minimize the ratio of cash holdings to corporate payments usually
require systematic analysis of cash flows. However, many opportunities for husbanding cash resources can be found without undertaking such studies. Frequently, opportunities for synchronizing the flow of receipts and payments are
quite obvious and cash needs can be reduced by making use of the funds in duplicate or little-used bank accounts. Many internal corporate policies have been

Chart8. MEASURES OF THE RELATIONSHIP BETWEEN MONEY PAYMENTS
AND CASH HOLDINGS OF NONFINANCIAL CORPORATIONS
Percent

Percent
Cash as a percentage\\
of gross sales
\
-*----- Scale

/
..,"

•

/
/
i
i
I
I

...-"
re-

Nonfinancial transactions
velocity
Sca le--..-

4—

3L--

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

'
cl 1
1
1930

1 1
1 1
1935

1
1
1940

1 1 1
1 1
1945

1
1
1950

1 1
III!
1955
1960
1965
* Beginning in 1958,"business receipts" were substituted
for "gross sales" which are no longer reported.
1- New series--prepared by Paul F. McGouldrick,formerly of
the Boa rd of Governors of the Federal Reserve System-beginning in 1960.

Sources: Board of Governors of the Federal Reserve System
and the Internal Revenue Service.

developed which make executives conscious of the need to economize on cash
and aware of the opportunities for doing so.
Systematic analysis and the projection of corporation cash flows are not new
developments,' but their use has become generalized only since World War II.
An increasing number of corporations are now using cash budgets which, in some
cases, are prepared for several years in advance but are frequently reviewed and
revised for the near term.'Such budgeting involves relating cash needs to projected
levels of sales and capital expenditures under alternative sets of assumptions. Execution of a cash budget requires close centralized control over cash flows and
resources.
An important aspect of corporate cash management concerns efforts to reduce
float, especially mail float. One means of reducing mail float and the volume of
uncollected balances is the decentralization of remittance collections (the establishment of regional collection accounts reduces mailing time). Decentralization,
together with the operation of lockbox systems, has substantially reduced mail float
for large national corporations. The lockbox system involves the interception of
checks at United States post offices and their channeling into collection areas as near
to their point of origin as economically feasible. A firm that uses this system instructs its customers in a given area to mail remittances to a specific post office box
address.' The depository bank then makes frequent pickups from this lockbox.
Checks are immediately processed for collection, and duplicates (on microfilm
or detached stubs) are forwarded to the corporation's home office for accounting
purposes. An East Coast company,for example, whose billings in California would
normally result in remittances being sent east and then returned to California for
collection, could—by using the lockbox system—have their billings paid to a post

3Alfred Sloan, recounting his many years with General Motors, states that, in 1922, "We began
calculating a month ahead what our cash would be each day of the month.... Against this projected
curve we compared each day the corporation's actual cash balances.... By reducing our cash balances
in banks, this system enabled us to invest the excess cash, principally in short-term government
securities." My Years with General Motors (New York, 1963), page 123.
'For detailed discussion, see Robert W. Johnson, Financial Management (3rd ed.; Boston, 1966),
Chapter 5, and J. Robert Lindsay and Arnold W. Sametz, Financial Management: An Analytical
Approach (revised ed.: Homewood, Illinois, 1967), Chapters 8, 9, and 10. See also: Roger M.Pegram,
"How Banking Serves the National Account", The Conference Board Record (May 1968), Kalman
J. Cohen and Frederick G. Hammer, "Deposit Turnover, Innovations and Bank Profitability", The
Banker's Magazine (Spring 1968), and Paul S. Nadler, "Banks Speed Up Velocity of Money with
Innovations, Hurt Themselves", American Banker (March 19, 1963).
Wormally, only corporations with annual sales in excess of $25 million are likely to make use of
such a service.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

office box address in a California city, thus eliminating the check-mailing time from
the East Coast to the West Coast and back. In turn the regional depository bank
periodically (daily or when balances reach a specified level) transfers collected
funds by wire, according to the company's instructions.
The benefits derived from using "remittance" banking depend on the geographic
pattern of collections. For a firm with national sales, remittance banking may
reduce float on the average of perhaps three days, most of it through the reduction
of mail float. As a result, collected balances of corporations using remittance banking or similar arrangements are increased while checkbook balances of payers
remain unchanged and their bank balances are reduced more promptly than when
routine procedures are used.
Efforts to reduce mail and bank float have been aided by the gradual shortening
of Federal Reserve availability schedules from the early 1920's when the maximum
was eight days to a two-day maximum deferment schedule established in 1951. In
fact, the latter allows part of the funds collected through the Federal Reserve System to become available to banks before actual collection (this being the origin of
part of the Federal Reserve float which in recent years has averaged about $2 billion). Other efforts of the System to speed up collections include arranging for
armored car routes and for consolidated air shipment of checks.
Another means of reducing cash needs is the use of drafts instead of checks
drawn on banks. Though employed by a number of large companies for various
kinds of payments, the use of drafts is not very widespread. Funds to cover drafts
need not be available in the bank on which they are drawn until the drafts are
actually presented for payment, though checks should, of course, be drawn only
against "good" funds available at the time of drawing on the account. Balances
must be maintained to cover all checks issued, even though the issuer knows that
some of them will not be promptly presented for payment. By using drafts, however, companies can reduce their working balances to the level of expected draft
presentations.' The use of drafts by corporate treasurers also permits a greater
centralization of disbursement accounts, even when such drafts are made payable
"through" several banks strategically located throughout the country.

6A case in point is the use of drafts by the American Telephone and Telegraph Company when
paying its quarterly dividends. By studying the time pattern of collection of its dividend checks,
A T & T concluded that significant savings could be realized by substituting drafts for checks. Instead
of depositing with the paying bank an amount equal to all dividend checks, it now deposits each day
only an amount calculated (on the basis of a probability model) to cover the drafts expected for
presentation that day, retaining the use of funds for shareholders who delay cashing dividend drafts.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

INVESTING TEMPORARILY REDUNDANT CORPORATE CASH

Employment of temporarily redundant funds in the money market by corporations,
municipal governments,' and others has contributed to rising postwar income
velocity and, at the same time, has increased the transactions velocity of demand
deposits by generating an additional and frequently large flow of financial debits.
Corporate funds available for short-term investment usually belong to one of
the following broad categories: (a) accumulation of tax reserves, in particular for
the quarterly payment of corporate income taxes, (b) accumulation of funds for
the payment of dividends and interest and for debt repayment or amortization,
(c) funds temporarily available because of the seasonal nature of sales or of
certain categories of operating expenditures,(d) funds kept liquid to take advantage of attractive business opportunities (e.g., to exercise options, to make deposits
on bids), (e) proceeds of long-term financing held pending disbursement in connection with investment in plant and equipment, and finally (f) general liquidity
reserves, or what some analysts, following Keynes, refer to as balances held for
speculative or precautionary motives. In practice, funds that are accumulated for
specific use in the future are taken into account when decisions are made with
regard to the amount of liquid assets and cash that a corporation may want to hold
at any time to achieve the desired portfolio balance in its financial investments.
Alternative business uses of such balances may involve acquisition of nonmonetary
liquid assets, repayments of short-term borrowings, or extension of credit, mainly
in the form of trade credit.
The range of instruments in which temporarily redundant corporate cash may
be invested has become increasingly extensive (more fully discussed in Chapter 2).
Some treasurers have been successful in persuading their boards of directors to
lengthen the maximum maturity of approved media for the investment of temporarily surplus funds and also to lengthen the list to include even foreign securities and trade paper.

7There is considerable evidence that so far state and local governments have been lagging in
introducing scientific principles of cash management. See J. R. Aronson, "The Idle Cash Balance of
State and Local Governments: An Economic Problem of National Concern", Journal of Finance
(June 1968), pages 499-508.
Aronson's estimates, derived from an inventory approach model, are based on a single year's
(1962) data, aggregate state and local government units, and have other limitations. His conclusion
that in that year two thirds or more of cash balances actually held were in excess of optimal balances
should be considered as merely suggestive of the magnitudes involved for this segment which, according to flow-of-funds estimates, in recent years held about 4 to 6 percent of total cash balances.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Government securities dealers, finance companies, and other users of short-term
funds have greatly contributed to spreading the knowledge, and consequently the
use, of money market instruments. Thus,the widening employment of repurchase
agreements has to some extent resulted from the systematic efforts of users to
seek out corporations likely to be, at least intermittently, a source of money market
funds. The banks, too, have promoted more sophisticated liquidity management.
Indeed, banks have in some instances assumed the function of managing investment of excess corporate cash. In effect, by competing for customers in offering
cash-economizing or liquidity-management services, banks tend to reduce the
level of aggregate business deposits in relation to a given volume of business payments for the banking system as a whole.
On the other hand, some corporate treasurers are now taking the view that income obtainable from the investment of surplus funds and from the reduction of
cash balances in general is not worth the extra effort required, and that such policies
entail a risk of losing the goodwill of bankers which they may need in periods of
credit shortage. The difficulties experienced by some corporations in obtaining
bank loans during the stringent credit conditions of mid-1966 may have strengthened the interest of corporations in maintaining good bank relations.
The various means available to business firms, especially to national corporations, for reducing cash holdings in relation to payments and for investing temporarily redundant funds are in a large way responsible for the increased turnover
rates of corporate balances (and thus, of total balances) since World War II. Due
to the familiar ratchet effect, these institutional and structural developments, and
the initiating forces that explain their adoption, tend to increase velocity in the
long run as well as cyclically. A process put in motion by a given cause (e.g., high
or rising interest rates) is not reversed once the cause disappears or its intensity
diminishes. Rising money rates lead to economizing on balances but, once new
techniques are developed to achieve the desired results, they become a more or less
permanent feature of payment and cash management techniques.'

8That the various "near moneys" have acted as money substitutes for the corporate sector can be
inferred from the fact that, in about two thirds of the quarters between 1952 and mid-1966, corporate
money holdings moved in an opposite direction from changes in holdings of other financial assets
(Government securities, time deposits, and finance paper). See Gloria Shatto, "Money Substitutes
for the Corporate Business Sector", Journal of Finance (March 1967), page 84. For a very different
interpretation, see Milton Friedman and Anna J. Schwartz, A Monetary History of the United States,
1867-1960(Princeton, New Jersey, 1963)pages 659-72.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

ECONOMIZING ON PERSONAL BALANCES

Longer term changes in the velocity of consumer balances reflect a variety of influences, such as: the tendency to minimize checking account activity in order to reduce service charges, the more widespread use of charge accounts and other forms
of consumer credit, the recent growth of credit-card and overdraft arrangements,
and the increasing attractiveness of various kinds of thrift accounts, and more
recently of consumer certificates of deposit. Other developments include withholding of income taxes, union dues, payments into welfare funds, subscriptions
to Blue Cross and similar plans, and deduction for systematic savings by employers.
No-minimum-balance accounts and the widening practice (facilitated by the use
of computers) of paying interest on savings accounts encourage depositors to reduce demand balances to a minimum. Protection offered against checks being returned for insufficient funds by automatic crediting of a depositor's account with
a loan (in fact, granting overdraft facilities) or transferring of funds automatically
from his savings account under prenegotiated arrangements has the same effect.
The most important influence affecting consumer money holdings since World
War II is the use of credit by households. Since the 1930's, banks and other lenders
have widened the range of credit facilities available to individuals for personal
expenditures. The rise in per capita income certainly reinforced this trend, but
perhaps even more important was the quest by banks for new outlets for funds to
replace short-term business loans, which had declined sharply in importance during
and after the Great Depression as a consequence of the more extensive use by
corporations of internally generated funds. In response to these stimuli, the banks
have increasingly advertised their consumer credit activities and have competed
quite successfully with the financial institutions which specialize in this field.
The increasingly rapid proliferation of charge accounts and general credit cards
has, of course, tended to encourage the use of checks as a means of settlement as
well as to bring payment dates closer to income receipt dates and therefore to
reduce the need for maintaining cash balances in the intermediate period. After
the end of World War II, the use of credit cards issued by the major oil companies
and hotel chains became fairly widespread. More recently, this form of credit has
rapidly expanded into general-purpose credit cards which permit charging a very
wide range of goods and services, thereby making such cards the equivalent of
cash. Nor is the use of credit cards limited to United States territory; credit cards
are now available for consumers to charge the purchase of goods and services
while traveling abroad. Carrying a credit card—or, as is often the case, a whole
collection of credit cards—makes it unnecessary for the individual to have large
76

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

amounts of cash in his possession. Nor does he have to have the resource
s, presently or in the immediate future, to cover purchases;some bank credit cards
permit
the holder to "charge" a loan (equivalent to an overdraft privilege),
thereby
further reducing the need for cash.'
Indeed, more recently, a number of banks have moved in the direction of
extending what amounts to limited overdraft facilities by giving custome
rs the privilege
of drawing checks without adequate covering balances. Under
each of the several
variants of such check credit plans, loan balances are automatically establis
hed
(subject to some stipulated ceiling) when checks are presented in excess
of available "good" balances, which are then subsequently amortized by periodic
payments, like regular personal loans. Giving consumers access to "instant"
money
for any purpose tends to reduce demand for liquidity and deposit balance
s."
Just as bank services in the personal credit field have expanded rapidly
since
the war, so have other sources of funds for emergencies, such as those offered
by
personal finance companies, credit unions, and union and company welfare
funds.
Medicare, Blue Cross, and other medical plans meanwhile have lessened the
financial impact of sudden illness. Easy access to funds to meet unforeseen expense
s
and the wider availability and use of various forms of insurance to meet
them
reduce the need for individuals to maintain cash balances for these purpose
s.
The greater use of credit by consumers and the availability of emergency services on a prepaid or insurance basis have tended to reduce the need for liquidit
y.
Unemployment insurance and other forms of social security, company supplemental unemployment benefits and retirement plans, as well as the increase in job
security as a result of collective bargaining, including the spread of seniority rules,
have considerably reduced the need for protection against sudden changes
in
income. More importantly, the absence of any severe and prolonged depression
since World War II has encouraged consumer readiness to buy on credit, and thus
has tended to weaken the link between cash balances and the level of spending.

90ne New York bank has named its credit card "The Everything Card".
charged to credit cards are business rather than personal expense; this is To some extent, payments
especially true of business
entertainment and travel expenditures.
Credit extended through bank credit card and related (including check credit)
1967 amounted to only $198 million, according to Bank Credit-Card and Check-Cr plans in September
edit Plans (a report
of the Board of Governors of the Federal Reserve System, Washington, D.C.,
July 1968), page 3. The
development of this technique is, however, still in its early stages. In
contrast,
nonbank
credit cards,
for which comparable data are not available, generate a much smaller
amount of debits and credits.
"Since the account is usually credited simultaneously only with the
amount required to cover the
check, this arrangement increases Vt as well as V.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

7. Implications of Recent Changes in Velocity
The sum of the reactions of all economic units to changes in the money supply
determines whether, when, and to what extent these variations cause aggregate
spending to change rather than be absorbed (or reinforced) by variations in velocity.
Thus, velocity reflects the economy's reaction to changes in the money supply induced by the monetary authorities, together with influences traceable to largely
technical factors.
Velocity is not constant over time, as assumed in the older version of the
quantity theory of money, including that formulated in the Cambridge equation;
nor is it a stable function of "permanent income" or wealth alone, or so strongly
dependent on one single determinant, such as interest rates, as to make possible
firm projections of its behavior in the long run, and certainly not in the short run.
Variability in velocity breaks the rigid link between money and income, since
changes in the money supply, however induced, may result in pushing velocity up
or down rather than produce the desired effects on spending and income. Contemporary advocates of a monetary policy which would concentrate on the control
of the money supply do not endorse the view of the older "crude" quantity theory
of money which assumed constant velocity. But they must assume that the link
between the quantity of money and the level of economic activity is stable and predictable. Thus, much effort has been expended, particularly in the period since the
appearance of the earlier version of this study, in developing theories and constructing models that would establish such stable relationships with identifiable
underlying forces. Researchers have also examined the behavior of velocity since
the end of World War II to ascertain whether the observed rise was due to some
exceptional or transitory forces which would soon spend themselves and permit
velocity to level off. In particular, the question was raised whether some kind of
"velocity ceiling" existed which would not likely be exceeded, at least in the foreseeable future. To some extent, views on the outlook for velocity have changed as
the rise in velocity has shown no sign of abating.
78

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

As turnover rates approached the levels established in the twentie
s, when they
were influenced by the exceptionally large volume of stock market
transactions, it
seemed logical to conclude that recent increases in velocity experi
enced in the
1950's were not sustainable. Writing in 1959, Professor Lawren
ce S. Ritter expressed the belief "that present limits to [income] velocity are closer
to an annual
rate of about 31
/
2 rather than the peak of 4 times per year reached in the 1920's".1
By 1963, with income velocity approaching a turnover rate of 4
times per annum,
and showing no signs of slowing down, he observed that: "There
is thus less evidence today that velocity is approaching a ceiling than there was six
years ago ....
there is probably considerable room for further advance still remain
ing."2 By
1967, when velocity had surpassed the 1929 peak and stood at
around 4½,Professor Ritter was willing to speculate about the probability of
a velocity rate of 7
or more in 1986 and chided those who still believed that "busine
ss firms and
consumers have already learned just about all there is to know about
how to economize on cash balances, and that no financial innovations are
likely to arise over
the coming decades which will make cash conservation
even more possible than
it is today"!
The post-World War II behavior of velocity has also cast
doubt on the validity
of the explanation advanced by Milton Friedman and Anna
Schwartz in what has
been perhaps the most ambitious attempt to explain velocit
y changes in the framework of a broad theory of economic behavior. Friedman and Schwar
tz regard
money merely as a "temporary abode of purchasing power"'rather
than as fulfilling several functions that can be meaningfully distinguished for analyti
cal purposes, as Keynes and others did. Thus, in linking the demand for money
uniquely
to the level of "permanent income"(measured to eliminate temporary or
transitory
influences), they are eclectic with regard to the proper definition of money,
recog-

1Lawrence S. Ritter, "Income Velocity and Anti-Inflationary
Monetary Policy", American Economic Review (March 1959), page 125. A little later, Professo
"... it seems unlikely that aggregate velocity will long continue r Richard T. Selden concluded that
to rise anywhere near its recent rate,
and a resumption of the prewar secular decline would not
be at all surprising". "The Postwar Rise
in the Velocity of Money: A Sectoral Analysis", Journal of Finance
(December 1961), page 533.
2Lawrence S. Ritter, "The Role of Money in Keynesian
Theory"
, Banking and Monetary Studies,
ed. Deane Carson (Homewood, Illinois, 1963), page 150.
3Lawrence S. Ritter, "How Fast Does Money Run?" Economi
c Report (Manufacturers Hanover
Trust Company, March 1967).
4Milton Friedman and Anna J. Schwartz,
(Princeton, New Jersey, 1963), pages 649-50. A Monetary History of the United States, 1867-1960

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

s may fall
nizing that a number of claims besides currency and demand deposit
however, they
within this interpretation. Primarily for reasons of convenience,
cial bank
commer
time)
plus
d
choose to define money as currency and total(deman
is that
choice
deposits held by the public.' A major consideration underlying this
d on a regular
prior to 1914 demand and time deposits were not separately reporte
and consistent basis.
long-run beFriedman and Schwartz have advanced the proposition that the
"luxury good", the
havior of velocity is explained by the position of money as a
. The logic
demand for which rises more than in proportion to increases in income
they
wealth
their
e
increas
units
of their view is simple and appealing: as economic
of
form
the
in
can better afford the luxury of holding a larger portion of wealth
decline over
liquidity balances. As a consequence, velocity may be expected to
Schwartz to
and
an
s
Friedm
compel
the long run. At the same time, such a claim
to decline.
failed
search for special explanations to apply to periods when velocity
of the
The data presented by Friedman and Schwartz show that income velocity
1.7 in
broadly defined money supply declined from about 4.6 in 1869 to about
in vedecline
n
long-ru
the
that
1960. Nevertheless, it is obvious from their data
World
of
ng
locity occurred primarily in the years between 1880 and the beginni
speculate
War 1.6 For the years before 1880 no trend is visible, though one could
apparent
any
there
is
nor
d,
decline
for any number of reasons that it should have
readily
is
1946
and
trend for the period 1914 to 1929. The decline between 1929
without
accounted for by the unusual circumstances of the depression and the war,
recourse to the "luxury goods" hypothesis.
related
While the broad decline in velocity between 1880 and 1914 is no doubt
least
at
played
to the rise in per capita income, it is likely that institutional forces
l for
as important a role. Indeed, it is possible that the rise in income is but a coveral
in
perhaps
and
period,
this
in
ted
interac
a number of institutional factors which
period
the
of
aspect
One
earlier times as well, to increase the demand for money.
that the
from 1880 to about 1914 is of special interest. It was during these years
' The expancommercial banking industry experienced especially rapid growth.
d subdecline
which
y,
currenc
of
sion of deposits came partly at the expense

5Ibid., page 649,footnote 9.
'Ibid., Chart 57, page 640.
. "The Monetary Inter7James Tobin termed these years "the great day for commercial banking"
(June 1965), page 475.
pretation of History, A Review Article", American Economic Review

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

stantially in relation to total deposits. But perhaps equally significant the growth
of commercial bank deposits coincided with the sharp relative decline of mutual
savings banks as repositories of personal savings in the country as a whole even
though they continued to be important repositories of financial savings in the states
where they were operating. In the five years preceding 1880, mutual savings bank
deposits were equal to about 70 percent of all commercial bank deposits, but by
the turn of the century this share had declined to 25 percent. In great part, the
relative decline of mutual savings banks (in this period) occurred because they
were confined essentially to the older and slower growing North Atlantic states,
but even in those states their position tended to be eroded. For the country as a
whole, the inflow of savings deposits into the commercial banking system must
have been very large (especially in the case of "country" banks and banks away
from the Eastern Seaboard), even though the large New York City banks manifested no interest in attracting or accommodating savings deposits almost up to
the beginning of World War I. As a result, at least some of the decline in velocity
between 1880 and 1914 is attributable to the growing role of commercial banks
as thrift institutions.
Other institutional factors operated also in the direction of causing velocity to
decline. Among these other influences was the growing share of output that
passed through the market place, thus increasing the need for transactions balances.
The enormous expansion in the number of banking offices, particularly in the Midwest and.South, may have had a similar effect. Most important, however, were
the changes taking place as a consequence of increasing urbanization and the
growing significance of commerce and industry in the uses of credit. Especially for
households, purchase by credit was far more important in this period (and more
so in the early part of the period) than in subsequent years, at least until the
decades since 1920. In agricultural regions, it was not uncommon for money to be
received perhaps only once or twice a year and, then, almost immediately paid out
to retire debts accumulated over the previous months. Though such practices had,
of course, been more widespread in earlier periods of American history, they undoubtedly still characterized the way in which large numbers of economic units
(e.g., farmers and country merchants) operated. in the newer regions of the
1880's and 1890's, where agricultural and/or mining activities dominated and
where banks were relatively few, or where their credits were largely of nonmercantile sorts, it should not surprise us that velocity was quite high. Nor, if this line
of reasoning is valid, need we invoke the "luxury" hypothesis to explain the subsequent decline in velocity as the nation phased out its remaining frontier regions

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

and settled down to the building of an urban and industrial society.'
The post-World War II rise in velocity, conflicting as it does with their "luxury
goods" hypothesis, has caused Friedman and Schwartz to advance an explanation
which is based on an important aspect of changes occurring in our economy since
the thirties: "Other things being the same, it is highly plausible that the fraction of
their assets individuals and business enterprises wish to hold in the form of money,
and also in the form of close substitutes for money, will be smaller when they look
forward to a period of stable economic conditions than when they anticipate disturbed and uncertain conditions."' Once adjustment to a stable framework of expectations is completed after war-generated disturbances, velocity will again rise."
There is, however, no need to search for special explanations. Whereas between
about 1880 and 1914 commercial bank deposits grew at the expense of mutual
savings bank deposits, in the fifties and early sixties the thrift "deposits" of savings
and loan associations and credit unions were growing at the expense of commercial
bank deposits: this time, institutional change was working against the commercial
banks. If savings bank deposits and savings and loan shares are encompassed in
the definition of the money supply, velocity would have declined irregularly during
the entire period from 1951 on, as shown in Chart 3.
There is, indeed, a natural tendency when dealing with a long span of time to
concentrate on quantifiable influences (time series) and to disregard institutional
changes and influences which in various subperiods shape decisions and expectations. We can observe the importance of institutional factors when dealing with
the current scene; we have to reconstruct them for past periods and try to assess
their significance through statistical tests. In this process, we may miss some of
what was relevant at the time but which did not find proper sediment in statistical
data or even in contemporary published comment.
Econometric research into systematic relationships between money and other
relevant magnitudes is, by necessity, backward directed. It labors under the handicap of inadequate data and can allow for only a limited number of structural and

'Martin R. Blyn, "Income Velocity, 1880-1920: A Regional Analysis" (unpublished manuscript).
'Friedman and Schwartz, op. cit., page 673.
.one might suppose that by 1960 expectations were approaching a plateau. If this be so, if the
present interpretation is right, and if the experienced degree of economic stability shows no drastic
change, one might expect the rise in velocity to end and the long-term downward trend to emerge once
more." Friedman and Schwartz, op. cit., page 675. One might ask why the stability expectations of the
"new era" of the twenties did not produce a similar downward adjustment.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

institutional influences. A very large number of econometric studies on money
have been published in recent years; they have enlarged and organized our knowledge of relationships prevailing in past periods of various horizons. Most of the
results of the studies are plausible, stressing the role of interest rates and some other
variable(s), such as income and/or wealth. In fact, velocity (the demand for
money), interest rates, and income are all correlated with general business conditions, and in earlier business-cycle research (as mentioned in Chapter 5) velocity
was widely used as an indicator of the cyclical position of the economy. Supply
conditions are considered in demand-for-money models only indirectly, if at all, by
assuming that monetary policy is reflected in interest rates. Thus,the rate used may
be presumed to reflect current monetary and fiscal policies as well as expectations
with regard to them.
Students are far from being in agreement on the relevant variables for a demandfor-money function and on the stability of such a function over time. Alternative
rationalizations, however, can often be supported on the basis of statistical results
of substantially equal persuasiveness. The limits set for this publication would be
far exceeded even if an attempt were made to review the main econometric studies
on the demand for money and their assumptions and implications. This is a matter
of regret, tempered only by our conviction that available data are insufficient to
undertake analysis on a level of disaggregation by economic sectors which we
would accept as analytically meaningful."
Focusing analysis on the nexus between aggregate monetary measures of economic activity and money, i.e., velocity, rather than on the demand function of
money, forces the analyst to consider the institutional environment and other
influences that help to explain breaks in continuity or temporary deviations."

"Thus, for instance, a comparison of estimates based on household survey results and those prepared annually as part of the flow-of-funds estimates shows considerable differences for the crucial
household sector. For 1963, a year in which such a comparison can be made, the flow-of-funds tables
estimate demand deposits and currency holdings in this sector to have increased by almost twice the
amount estimated on the basis of a survey sponsored by the Board of Governors ($4.3 billion versus
$2.2 billion), while for total savings accounts the difference was even greater ($23.0 billion versus
$10.8 billion). For the two years 1960 and 1961, the flow-of-funds figures indicate an increase for
the
combined holdings of money and savings accounts averaging $14.8 billion, while a survey found an
increase amounting to less than half a billion dollars. See Dorothy S. Projector, Survey of Changes
in Family Finances (Washington, D.C.: Board of Governors of the Federal Reserve System, 1968)
Table 6. The fact that in the flow-of-funds accounts the household sector includes nonprofit institutions
cannot possibly account for even a small portion of this difference.
"Such as the effect of bottlenecks in the operations of major securities exchanges and the related
clearing of securities and payments for them in 1968 which was a significant influence in bulging the
money supply.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Indeed, recent developments in the payments mechanism and in the management
of liquidity strongly suggest that we are at the threshold of important new developments which are likely to bring further significant changes in the relationship of
money to the flow of payments. How soon these developments will alter the payments mechanism and the demand for money is still largely a matter of conjecture.
One does not have to expect, however, that our check-ridden society will shortly
be succeeded by the millennium of a checkless society: even less revolutionary
changes in banking are likely to have considerable implications for velocity. One
can, for instance, envisage the possibility of applying the overdrafts to business
accounts and more generally replacing compensating balances with explicit service
charges. But, if our basic payments mechanism shifts from the use of checks to
credit transfers employing electronic impulses transmitted instantaneously through
a system of interconnected electronic devices, much of what has been written about
the velocity of, and demand for, money will doubtless become obsolete, and the
very concept of money will undergo a radical change.
In the meantime other developments affecting demand and supply factors, as
well as institutional innovations relating to financial instruments and markets, are
continuously changing the position of money in relation to other instruments of
liquidity in the various sectors of the economy." Some of these institutional factors
involve legislative developments, others represent the general evolution of needs
and the market response to them. The introduction of deposit insurance at
commercial banks in 1933 was such a structural change, and its significance increased over time, as similar provisions were passed to insure deposits at mutual
savings banks and at savings and loan associations and as the maximum limits
of insurable funds were gradually raised. Regulation Q and related rulings by
other supervisory authorities limiting maximum allowable interest rates provided
the basis for administrative actions which from time to time have changed the
competitive position of individual substitutes for money as instruments of liquidity.
Some other changes resulted from competitive actions taken by individual banks

"Professor L. Ritter, after analyzing some of these changes, has concluded that "Already the financial asset we call 'money' (currency and demand deposits) has lost ground in our evolving financial
system.... A larger and larger superstructure of debt, credit and economic activity is being erected
atop a relatively diminishing monetary base.
"What we are likely to see, then, over the coming decades, is a financial system in which techniques
of portfolio and liquidity management advance far beyond the present state of the art, in which competition intensifies between and among both borrowers and lenders, and in which 'money' as we know
it today starts to lose its present characteristics and takes on new ones in their place." Bank Liquidity
Reexamined (Association of Reserve City Bankers, New York, 1967), page 32.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

and other depository institutions which, by permitting withdrawal of saving
s
accounts at other-than-stated quarterly interest payment dates and by
introducing
daily payment of interest, increased the liquidity of savings deposit
s as well as the
returns available on them. The period encompassing the monetary policy
of wargenerated constraints through the Treasury-Federal Reserve "Acco
rd" of 1951,
the growing interdependence of the world's economies and financi market
al
s following the introduction of convertibility by the major countries in 1958,
the developing of trade in dollars and of the flotation of dollar securities abroad,
and finally
the unprecedented long period of domestic expansion since 1961 has
raised many,
still unanswered, questions as to how these events will affect interes
t rate levels
and thus the opportunity cost of holding money. We are more impres
sed by the
potential for future change in the demand for money in the contin
uously evolving
framework of institutional factors than by the consistency of
patterns derived
from time series for past periods.
The years after World War II have seen a shift in demand away
from money
into nonmonetary liquidity claims, as several new instruments
have come into
being to compete with the older instruments—such as United
States Government
securities (mainly Treasury bills), commercial paper, including
finance company
paper, and bankers' acceptances—while thrift institutions
and commercial banks
have made various types of time deposits more attractive
to households. In the
words of one researcher,"all of the elasticities [of money
demand, with respect to
income and interest rates] appear to have declined signifi
cantly in the postwar
period as compared to the prewar years"." Indeed, the shift
in asset-holder preference suggests not so much that cash and the near money
s are substitutes as that
money is becoming an "inferior" asset in relation to money
substitutes. Undoubtedly, rising interest rate levels provide much of the explanation
as to why money has
become an inferior asset, compared with the other liquidi
ty instruments. But it
is also clear that the far-reaching institutional changes that
have taken place in
corporate liquidity management since the late fifties
have been a very important

"Thus, whereas elasticities of —.1956 and .9344 were obtained
for interest rates and income, respectively, in the period 1924-41, elasticities for the period covering
fourth quarter of 1959 declined to —.0538 and .5130, Ronald the fourth quarter of 1946 through the
L. Teigen, "Demand and Supply Functions for Money in the United States: Some Structural
Estimates", Econometrica (October 1964),
pages 501-3. See also David Laidler, "The Rate of Interest
Empirical Evidence", The Journal of Political Economy (Decemband the Demand for Money—Some
er 1966). For the contrary view that
the demand for the money function has not changed, see
Allan H. Meltzer, "The Demand for Money:
The Evidence from the Time Series", The Journal of Political
Economy (June 1963).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

influence."
The introduction of the certificate of deposit (CD) in 1961, with the simultament
neous creation of a secondary market, is perhaps the most significant develop
tions
since World War II in influencing the cash management policies of corpora
municipal
and of other holders of large cash balances, including those of state and
ons.
instituti
t
nonprofi
governments, the pension funds which they sponsor, and
that
noted
When the first version of this monograph appeared in 1959, it was
thirties
corporations "make relatively little use of time deposits . . . . since the
28).
(page
nance"
predomi
of
Treasury bills ... have come to occupy a position
commer
many
Indeed, until that time corporate time deposits were unwanted at
an
cial banks. Corporations and others were reluctant to use for liquidity reserves
introThe
time.
of
period
fixed
a
arrangement under which funds were tied up for
te
duction of the negotiable CD, however, has changed the entire picture of corpora
holdliquidity management. Between 1959 and 1968, nonfinancial corporate cash
GovernStates
United
erm
short-t
billion,
ings declined from $32.6 billion to $28.7
increased
ment securities from $17.3 billion to $9.2 billion, while time deposits
from only $1.5 billion to $25.1 billion (end-of-year figures).
of negotiaOther financial innovations were less dramatic than the introduction
short-term
various
II,
War
ble CD's (see Table 3). Thus, since the end of World
y instruliquidit
securities of Government agencies were added to the range of
grew only
ments used by corporations and others. While the Federal debt itself
es issued
securiti
States
United
in
slowly, there occurred an almost fivefold increase
ment),
Govern
States
by Federal agencies (not directly guaranteed by the United
years later.
from $4 billion at the end of 1952 to more than $22 billion fifteen
ents
Although not direct obligations of the United States Government,these instrum
offer
thus
and
ment
Govern
are backed by the full faith and credit of the Federal
years, the
substantially the same security with appreciably higher yields. In recent
large.
proportion of such obligations maturing within one year has been very
market
active
an
which
(for
paper
ial
Among private obligations, commerc
in
growing
pace,
lar
existed in the years prior to 1929) increased at a spectacu
the
at
volume from a little over $3 billion at the end of 1957 to $20.5 billion
ance after World
end of 1968. Finance company paper especially gained in signific

paper rate in 1929 rose to almost
15Interest rates were also high in the twenties (the commercial
market "provided an equally
money
call
the
comment,
Schwartz
and
Friedman
as
and,
6 percent)
Friedman and Schwartz, op. cit.,
periods".
short
for
funds
investing
of
means
t
convenien
and
safe
page 660.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

TABLE 3

SELECTED SHORT-TERM DEBT OUTSTANDING*
Billions of dollars
December December December December December
Type of instrument
1950
1955
1960
1965
1966
Treasury bills
13.6
22.3
39.4
60.2
64.7
Other marketable Treasury
securities
38.9
38.3
34.4
33.2
40.5
Federal agency issues
1.1
2.1
4.4
9.2
13.4
Total Federal Government
and agency securities
53.6
62.7
78.2
102.6
118.6
Commercial paper
Bankers' acceptances
Negotiable certificates of
depositt
Total other short-term debt
Grand total

December
1968
75.0
33.6
13.3
121.9

0.3
0.4

2.0
0.6

4.5
2.0

9.1
3.4

13.3
3.6

20.5
4.4

0.7
54.3

2.6
65.3

6.5
84.7

16.1
28.6
131.2

15.7
32.6
151.2

22.8
47.7
169.6

*Debt maturing within one year; end-of-December figures.
fOutstandings for all weekly reporting banks on the last Wednes
day of December.
Source: Federal Reserve Bulletin.

War II. Commercial paper has an important advantage arising
from the ability
of the issuer to tailor the size and maturity of each obligation
to the specific
needs of corporations (or of other investors) employing temporarily
idle funds.
Very short maturities are available, and some issuers have arrang
ements under
which funds available for one day only can be invested in commer
cial paper.
Similarly, some corporations have utilized repurchase agreements
with Government securities dealers to obtain short-term investments that would
fit their needs.
In more recent years, corporations have extended liquidity manage
ment to the
international money market, especially by the buying and selling of
Euro-dollar
deposits as well as other currencies."
The decline of "redundant" uninvested funds, as accumulations of funds
to
meet scheduled payments and various reserves are invested (and continu
ously

"See Fred H. Klopstock, The Euro-dollar Market: Some Unresolv
ed Issues (Princeton, New
Jersey, 1968), in particular page 9 if.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

reinvested) to keep a maximum of such funds in income-yielding forms, has tended
the
to increase the turnover velocity of demand deposit accounts. A large part of
has
"idle"
as
to
funds which in earlier literature were not improperly referred
reinshifted into the high-turnover category, being continuously invested and
money
the
into
them
return
to
need
actual
vested in the money market until the
supply immediately prior to the date of disbursement arises.
Similarly, the range of financial assets which households in general consider
as per
practicable alternatives to holding money has widened since World War II,
dented
unprece
to
risen
has
capita personal income and the backlog of savings
has
levels. The resulting diversification of household holdings of financial assets
supplyalso been influenced significantly by the growth of financial intermediaries
ve
alternati
an
as
using
been
have
ing deposit liabilities which many households
increashave
ons
for, or as a complement to, a checking account. These instituti
bankingly offered many of the services traditionally associated with commercial
range
the
enlarge
to
them
enabling
on
ing and lately have striven to obtain legislati
in
iaries
intermed
of such services. The relative growth of these nonbank financial
istic
the 1950's and early 1960's was not a cyclical phenomenon; it was a character
claims
in
increase
The
alike."
ns
expansio
of the entire period, recessions and
nt in
issued by financial intermediaries, to individuals particularly, is importa
II.
explaining the rising income velocity of money since World War
result of
The rapid growth of thrift institutions since World War II was the
1950's,
the
out
Through
"
policies.
several factors, including aggressive advertising
rates
the
d
exceede
and early 1960's, the rates paid to depositors substantially
0's,
commercial banks were paying on time and savings deposits; after the mid-195
that
factors
ive
Qualitat
d.
narrowe
rates
however, the differences between these
acceptmore
them
made
enhanced the safety of savings and loan shares, and thus
e proable to the public, were also important." For instance, in 1950 the insuranc
so
visions of the Federal Savings and Loan Insurance Corporation were revised,

Controls", Banking and
"David I. Fand, "Intermediary Claims and the Adequacy of Our Monetary
Monetary Studies, ed. Deane Carson, pages 245-46.
ion Analysis (Englewood
"Edgar L. Feige, The Demand for Liquid Assets: A Temporal Cross-Sect
s were mainly involved in financCliffs, New Jersey, 1964), page 19. The fact that the savings institution
return were relatively high,
ing the booming residential housing market of the 1950's, where net rates of
payment of high returns.
must also be included in any explanation—at least as a factor permitting the
Money", Journal of
for
Liabilities
ary
"Tong Hun Lee, "Substitutability of Nonbank Intermedi
Finance (September 1966).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

that depositors could recover a larger portion of their funds in the event of default
of an insured savings and loan institution. Also, the number of savings and loan
associations covered by Federal insurance increased substantially. Aware that
their attractiveness depends to an appreciable extent upon their ability to endow
their liabilities with the characteristics of near moneys, savings institutions normally waive any legal requirements that may exist with respect to prior withdrawal
notices, and they make passbook loans almost automatically.
It is a common characteristic of developed countries that the financial assets of
consumers rise more rapidly than income. While households tend to diversify their
holdings of financial assets, size distribution of income, convenience, limited information, and other factors tend to channel the bulk of consumer noncontractual
financial savings into intermediaries that insure deposit claims. The rapid postwar
rise of consumer holdings of savings deposits at commercial and mutual savings
banks and of "shares" in savings and loan associations (and even more the very
rapid growth of various types of savings and similar certificates on which higher
rates of return are permitted by rate regulations currently in effect) is not tantamount to an increase in the demand for money, particularly if money is regarded,
as it is by Milton Friedman, as a "temporary abode of purchasing power". Like
other types of financial investments, such deposits are held primarily to obtain a
return, not to meet a flow of payments obligations. Indeed, given our institutional
framework and prevailing patterns of household savings and financial behavior,
the bulk of consumer savings automatically results in additions to financial assets
which some analysts include with the money supply. As long as savings deposits
remain the preferred type of consumer financial savings, and the savings ratio
stays close to the level of recent years, savings deposits are likely to rise at a more
rapid rate than the money supply.
The very rapid growth in the fifties of nonbank intermediaries, both absolutely
and by comparison with the rather sluggish growth of commercial banks," gave rise

"Among the many articles and books on this subject are the following: J. G. Gurley and E. S. Shaw,
"Financial Intermediaries and the Savings-Investment Process", Journal of Finance (May 1956); J. G.
Gurley and E. S. Shaw, Money in a Theory of Finance (Washington, D. C., 1960); Warren L. Smith,
"Financial Intermediaries and Monetary Controls", Quarterly Journal of Economics (November
1959); David I. Fand, "Intermediary Claims and the Adequacy of Our Monetary Controls", op. cit.;
J. Aschheim, "Commercial Banks and Financial Intermediaries: Fallacies and Policy Implications",
The Journal of Political Economy (February 1959); and L. S. Ritter, "The Structure of Financial
Markets, Income Velocity and the Effectiveness of Monetary Policy", Schweizerische Zeitschrift für
Volkswirtschaft und Statistik (September 1962). For representative selections of papers on both sides
of the issue, see Chapters 12 and 13 in L. S. Ritter's Readings in Money and Banking, second ed.
(Boston, 1961).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

to a debate about the significance of this development for the effectiveness of monetary policy in promoting economic stability. The question was raised, in particular, as to whether in periods of expansion the shifting of funds by the public from
commercial banks (presumably from demand deposits) to intermediaries, in response to the higher interest rates offered by these institutions, offset at least partially the restrictive credit actions taken by the Federal Reserve System.
Consider a situation in which the monetary authorities decide that it is in the
public interest to pursue a restrictive credit policy. Suppose further that bank
reserves are just sufficient to support existing deposit liabilities and that considerations of liquidity do not permit an expansion of bank loans. Assume also that
rising mortgage rates and bond yields enable the various thrift institutions to raise
the rates paid on savings deposits. If individuals respond by shifting demand balances from commercial banks to, say, savings and loan associations, the narrowly
defined money supply is not affected. Balances owned by individuals are merely
transferred through the investment activities of savings institutions to builders,
other business firms, and governmental units. With additional construction activity
now made possible, the level of GNP will rise and, with the money supply unchanged, so will the rate of income (and of transactions) velocity. Similar consequences would follow a shift of deposits to savings and other time balances at
commercial banks, or to savings banks, or their use for the purchase of newly
issued securities or securities held by nonbank investors. The expansionary effects
depend upon the nature of the shifts (e.g., reduction in currency or demand balances by households or a shift of savings balances from commercial banks to thrift
institutions or into primary securities)."In principle, however, the various asset
substitutions mentioned all permit additional credit creation and income expansion. Efforts by the monetary authorities to limit the growth in the money supply
are thus thwarted to the extent velocity increases. The sharp increase in interest
rates (since 1965) has lessened this problem by putting the thrift institutions, with
their rather long-term asset structure and undifferentiated liabilities, at a competitive disadvantage vis-a-vis the commercial banks. At the same time, however, the

Donald
21.For more complete discussions, together with attempts at measuring these effects, see Finance
Shelby, "Some Implications of the Growth of Financial Intermediaries", Journal ofcit.;
D. I.
(December 1958); W. L. Smith, "Financial Intermediaries and Monetary Controls", op.
Jack M.
Fand, "Intermediary Claims and the Adequacy of Our Monetary Controls", op. cit.; and
Guttentag and Robert Lindsay, "The Uniqueness of Commercial Banks", The Journal of Political
Economy (September and October 1968), which contains a useful bibliography.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

rise in interest rates, to the highest levels in a century, has led (as in 1966 and
1969) to an increase in credit flows bypassing institutions (disintermediation)
which also tends to offset restrictive actions taken by the central bank.
There is no doubt that shifts of holder preferences between money and money
substitutes and the related activities of financial intermediaries that cause offsetting short-run movements in velocity have not infrequently tended to delay the
effectiveness of changes in monetary policy. It is now widely recognized that, in
countries with a developed money market, the money supply has two dimensions:
size and velocity. Fluctuations in velocity reflect and respond to changes in credit
conditions and monetary policy. Greater or less degrees of credit stringency also
tend to cause changes in the technical efficiency of money use. In a way, changes
in velocity signal to the monetary authorities how impersonal actions on their
part have affected the liquidity position of the economy as a whole, after the initial
reactions to the change in monetary policy have been worked out by the various
sectors and in the money-using processes.
Far from being inimical to the execution of monetary policy, changes in velocity
frequently perform a useful function. At times, the monetary authorities must
take massive action which may create temporary disturbances at the initial point
of impact. Usually there is a time lapse before policy actions in the open market
or before changes in the discount rate or reserve requirements permeate the entire
credit structure. In such a situation, velocity acts as a shock absorber and helps to
cushion and diffuse the initial effects of policy actions. To the extent that monetary authorities are aware of the nature and possible range of these chain reactions,
changes in velocity can be taken into account when determining the magnitude
and timing of a required policy action.
To sum up, the post-World War II rise in income (and transactions) velocity
was the result of two processes operating in the same direction. One was the
increased efficiency in the payment and clearing mechanisms, thus tending to
reduce transactions balance requirements in relation to a given volume of payments. The other was the gradual transfer of business and other nonhousehold
balances, excepting those required to support payments and credit, into money substitutes. The volume and diversification of money substitutes available to holders of
relatively large balances expanded in response to demand, and the performance of
financial markets improved in part as their size grew. At the same time, the attractiveness of savings accounts at commercial banks for temporarily redundant
funds of individuals grew. Other savings depositories aggressively sought to im:
prove their competitive positions, by offering depositors greater returns and by

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

making savings funds easily withdrawable whenever needed. One of the landmarks of the endeavors of depository institutions other than commercial banks to
persuade clients that their claims were liquid as well as profitably invested was the
success of savings and loan associations in obtaining legal sanction to designate
their liabilities as deposits rather than as shares (effective July 1, 1969).
Thus, since World War II funds not needed to perform a payments function—
either in the market for goods and services or to compensate commercial banks for
their services, including the provision of credit—were gradually shifted into incomeearning liquid assets. This process was stimulated by the rise in interest rates
that began after the "Accord" of 1951 and gained momentum during the long
and almost uninterrupted expansion of the economy that commenced ten years
later and -which is continuing at this writing. Certain events tended to hasten this
process, such as the introduction of marketable CD's and the offering of the
"magic fives" by the United States Treasury in October 1959, which dramatically
drew the attention of large groups of savers to the availability of higher returns
on easily accessible marketable instruments. The learning process is not even
over time, since it involves a large variety of units, each of which has a different
pattern of receipts and expenditures, temporary bulges of accumulated funds, and
varying liquidity needs. Moreover, the larger organizations tend to create special
units to manage cash flows and the cash position, so that returns may be maximized within the constraints of the particular liquidity needs of the unit. When the
current return on such operations becomes negligible or even less than the cost
of the special staff involved, cyclical declines in interest rates do not necessarily
result in their disbandment. But since the shift into money substitutes is encouraged
and facilitated by various specialized organizations originating such instruments or
providing markets for them, efforts to stimulate the use of near moneys are subject
to spurts when accelerated business activity leads to tight credit conditions and
associated higher interest rates. Of course, progress in technology underlying the
several aspects of improvements in the payments mechanism is also uneven. Iimovation in management is not more continuous, or more evenly spread over time,
than in technology.
For a number of reasons, judgments about the outlook for velocity are difficult
to make as the sixties draw to an end. Up to the present, the time required to process checks and the speed of transportation tended to set a definitive limit upon
further potential increases in velocity, while the availability of collected balances
was prerequisite for making good on a check issued. New influences are likely to
become more and more important in determining the relationship between money
92

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

balances and the flow of payments. One is the exploration of the feasibility of a
system of instantaneous payments through the use of electric impulses. Such
a
large-scale, and ultimately nationwide, computerized system would revolutionize
the entire payments mechanism by replacing the handling of checks with
transmission signals that would activate electronic book-entry systems. A checkles
s, or
rather less check-based, system would presumably involve some automatic
extension of credit in the form of limited overdraft facilities. Under such conditio
ns,
the need to have enough cash in the bank, on which current management of
cash
flows and cash budgeting is based, would lose much of its meaning. The amount
of
money underlying that part of the payments flows which would be cleared by
the
new system presumably would be considerably smaller than that required to make
the same amount of payments by check or currency, and the velocity of money will
increase greatly. It is also conceivable that the wider use of overdrafts, only recently
introduced for the household sector in the form of "ready cash" and similar plans,
could spread to business accounts even before the advent of the less-cash society.
"
The full potential of such truly revolutionary changes in the payments mechanism and the effect of the changes contemplated beyond the payments mechanism
are still difficult to visualize. But some experts envisage the advent of a credit
transfer system in which business firms and even households will arrange for
scheduled payments from their accounts to be transferred through an automated
clearings system by advising their bank of the amounts to pay, to whom,and when.
One can even go one step further:
Could the system be programmed so that for a given day's work no one
had more or less in his account, taking into account inflows, than
needed to cover outflows? If so, velocity would approach infinity and
money supply zero. However, in any foreseeable system we would
have a few billions of coin and currency and probably of demand deposits. What seems more to the point for the money manager is that no

22In a study based on data from several foreign countries in
which business-account overdrafts are
important ("Interpreting Monetary Statistics when Overdrafti
ng is Prevalent", an unpublished paper),
W. H. White has concluded that inclusion of unused overdraft
effect on calculations of transactions velocity (less than I percent facilities would have a negligible
of several years during the late forties and early fifties). Quoted for an average for various periods
in Anand G. Chandavarkar, "Unused
Bank Overdrafts: Their Implications for Monetary Analysis
and Policy", IMF Staff Papers (November 1968), which also contains a useful bibliography on
is by no means certain that this conclusion would be valid unused overdrafts in various countries. It
became widely available to business as well as to householdfor the United States if overdraft facilities
s.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

one will be holding money as a storehouse of value, a liquidity hoard or
anything other than a medium of exchange. Thus, the separation of the
liquidity and transaction functions of money may well force money
managers to turn themselves into liquidity managers, or into nearmoney managers,so to speak, and entail the development of an entirely
new set of tools for using the financial system to influence the economy's
growth and stability. The very least that will happen to us as money
managers will be some enforced rethinking of money supply theories in
a computerized economy."
Indeed,models and parameters derived from past experience would then become
clearly useless in predicting the future behavior of money balances. The indirect
effects of these contemplated changes in the payments mechanism on banking
activity, credit, and the entire financial system could be significant and perhaps
revolutionary. Nevertheless, we are still too far removed from the implementation
of any such system for its implications on velocity and on the role of money in relation to other instruments of liquidity to be explored in a meaningful way.

23Tomorrow's Money as Seen Today, remarks by George W. Mitchell, Member, Board of Governors
of
of the Federal Reserve System, at the annual stockholders meeting of the Federal Reserve Bank
Boston, Boston, Massachusetts, October 6, 1966.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

APPENDIXES

Contents:

Page
APPENDIX I: Composition of the Money Supply

Currency
Notes lost or destroyed
Currency carried or circulating abroad
Currency in effect circulating in the United States

98
99
99
100

Demand Deposits
Foreign-owned dollars not affecting GNP directly
Elimination of United States Treasury balances
Float adjustment
Certified and officers' checks
Structure of the money supply

101
101
103
104
105
107

What Deposit Balances?

108

APPENDIX II: Share of Currency in the Money Supply

113

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Appendix I:
Composition of the Money Supply
The purpose of this appendix is to examine the accuracy of the private money
supply series as a measure of the means of payment available to United States
residents and to those foreigners who use dollars primarily to purchase goods,
services, or investments in the United States market. What appears to be a clear-cut
statistical concept involves,in fact, numerous measuring problems,including fairly
arbitrary decisions about the allocations of certain categories of items of a mixed
character. Some broader questions, such as the relationship of money to other
categories of instruments of liquidity, have been discussed in Chapters 2 and 7.
Differences of view among economists exist as to what constitutes the money
supply. Some favor the narrow definition underlying the series currently published
in the Federal Reserve Bulletin which, in fact, covers deposits held by the domestic private sectors, by state and local governments, and by all foreigners;' others
prefer a broader concept to include, at least, commercial bank time deposits.
It should be noted that, prior to the creation of the Federal Reserve System,
differentiated reserve requirements in favor of time deposits did not exist and
many banks did not report such deposits separately. Thus, for that period, quite
arbitrary assumptions must be made in estimating the proportion of total deposits
represented by demand deposits. Chapter 4 discusses this issue and also comments
on the shortcomings of using the money supply which excludes United States
Government deposits to compute total income velocity and in using gross deposits
to compute transactions velocity.
Other issues of a technical nature are given little, if any, attention in academic
literature, as it is more concerned with the functions that money performs and with
demand-for-money models than with the underlying monetary statistics.' Little

iSee "A New Measure of the Money Supply", Federal Reserve Bulletin (October 1960), pages 110223; "Revision of Money Supply Series", Federal Reserve Bulletin (August 1962), pages 941-51; and
"Section I: Banks and the Monetary System", Supplement to Banking & Monetary Statistics (Washington, D. C., 1962). In these sources, a discussion of slight differences between this and the end-ofthe-month series may be found.
'Indeed, hardly any detailed independent investigation has been published since Lauchlin Currie's
pioneering study, The Supply and Control of Money in the United States (Cambridge, Massachusetts,
1934), in particular Chapter 3 which served as the basis for the first money supply series published
by the Board of Governors of the Federal Reserve System. See, however, a forthcoming monograph
by Milton Friedman and Anna J. Schwartz.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

attention has been paid outside the Federal Reserve System to the technical problems of measuring the money supply, and the descriptive material dealing with
the various revisions this series has undergone over time is almost entirely confined to articles in Federal Reserve publications. Our concern here, however, is
with those issues only that are directly relevant to the interpretation of velocity
statistics derived from money supply data.
Currency outside banks—one of the two components of the money supply—has
accounted since World War II for between 20 percent and 24 percent of the total
money supply, as published in the Federal Reserve Bulletin. (However, in most
of the years between the two world wars it accounted for a smaller proportion.)
Measuring this component involves a minimum of adjustment and no timing problems. In contrast, estimating private demand deposits involves a whole set of
assumptions as to timing, clearing procedures, accounting techniques, and the share
of checks chargeable to private deposits in the various components of float. Moreover, adjustments in the currency component affect primarily its level rather than
month-to-month changes; it is not certain that the same is true for the deposit
component. Deductions made from gross demand deposits to derive adjusted private demand deposits are quite large. For example, at the end of 1968, deductions from gross demand deposits amounted to $65 billion, or close to 30 percent.
Furthermore, the various adjustment items are subject to significant influences,
some of which are systematic but others are erratic. Estimating or reporting errors
in individual deduction items may have important effects on month-to-month
changes in the residual which is a measure of the demand deposit component of
the money supply series.

CURRENCY

The currency component of the private money supply is, in effect, the cumulative
total of all paper money issued and not demonetized or recorded as retired from
circulation (unfit notes) plus the domestic coins in circulation in the United States
as estimated by the Director of the Mint. This total is adjusted for amounts held in
bank vaults (including estimated amounts for nonmember banks). Account is
not taken of notes lost or destroyed (in disasters or inadvertently), circulating
abroad, or included in numismatic collections. The estimate of the stock of coin is
based on United States Treasury estimates of coin manufactured, damaged and
worn coins withdrawn from circulation, certain exports and imports, and general loss.
98

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Although the proportion of coins to the total amount of
currency in circulation has increased from 8 percent at the end of 1939 to 11 percent at the end of 1968, Federal Reserve notes constitute the bulk of our circulating
medium.' A relatively small part of currency outside the Treasury and Federal
Reserve Banks consists of notes in denominations of $500 or more, and there is
good reason to believe that these large-denomination notes are not, in fact, used
for day-by-day transactions.' The number of these notes has declined, as has their
share in the total dollar amount of notes outside the Treasury and Federal Reserve
Banks (from nearly 10 percent in 1939 to only 1.2 percent at the end of 1968).
Very large denomination notes are accumulated primarily for hoarding and tax
evasion purposes, here and abroad.
Available evidence suggests that the amount of currency lost and destroyed is
small enough to be disregarded for most analytical purposes. Valid estimates of
currency destroyed or lost can be made only for the several categories of paper
currency completely retired from circulation; amounts redeemed have invariably
fallen short of those shown as outstanding on Treasury or Federal Reserve System
records. Experience with the retirement of paper currency issued prior to 1929
(including gold certificates) indicates that approximately /
1 4 percent of all notes
issued was never presented for redemption.
NOTES LOST OR DESTROYED.

about one hundred years, United
States currency has circulated in several foreign countries, for various periods, on
almost equal footing with national notes and coins.' For example, in Cuba and
Haiti, United States currency at times represented a considerable proportion of
the circulating medium. Currently, Panama and Liberia use United States dollars
and do not issue any paper currency of their own. Since data are not available to
estimate the amount of dollars which circulate in several foreign countries along
with or in lieu of the national currencies, no correction can be made in United
States money supply data. The amounts involved are much diminished since World
War II and must surely be very small in relation to circulation in the United States.

CURRENCY CARRIED OR CIRCULATING ABROAD. For

3Including vault cash, which is not part of the money supply but which cannot be broken down
into notes and coin.
4Federal Reserve Bank shipments and receipts of such notes are very small, and notes are frequently returned in near-mint condition. In fact, no notes over $100 have been printed since 1945.
'See,for instance, John P. Young, Central American Currency and Finance (Princeton, New Jersey,
1925).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

United States currency is also carried by travelers into Canada, Mexico,Europe,
and other continents. While the bulk of both these flows is returned to the United
States (by banks which specialize in exchanging dollars or by returning travelers
or military personnel), the supply available domestically is reduced by amounts
that are unknown and that are subject to seasonal influences. Also, there is some
leakage into foreign hoards, particularly to countries with unstable currencies
and/or exchange controls.' Indeed, in some instances, United States currency has
been shipped to various countries to fill insistent demands arising, in part at least,
from the dollar's status as a preferred means of hoarding and of evading taxes
and foreign exchange controls.'
CURRENCY IN EFFECT CIRCULATING IN THE UNITED STATES. We can

summarize the dis-

cussion so far by the following equation:
Cr = Cr — (L + A)
C. =
Cr =
L =
A =

currency in effect circulating in the United States.
currency as shown in money supply statistics.
currency lost or destroyed.
currency circulating or hoarded abroad.

Currency in effect circulating in the United States consists primarily of currency
in the hands of consumers and business. The bulk of business currency is presumably held by retail trade and service establishments and by employers who
make wage payments in cash. It also includes currency held by the United States
Government, consisting of amounts held by post offices and paymasters of the
armed services. This treatment is contrary to that of demand deposits in the private
money supply series which excludes demand deposits held by the United States
Government (except for some disbursement accounts in remote locations of very
slight importance). Since no estimates of currency destroyed or lost, or held by the
United States Government, are available, and even a broader guess on"A"cannot
be made,no appropriate corrections can be made in the currency component of the
private money supply series, which thus contains a cumulative overstatement of
unknown magnitude (for further discussion, see Appendix II).

to pay the
6In periods of acute local inflation, American and other foreign firms may use dollars
salaries of their local employees.
currency
7In a recent two-year period, for instance, Brazil received shipments of United States
considerably in excess of $300 million.

100

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEMAND DEPOSITS

At least two problems are involved in interpreting the deposit component of the
money supply. One concerns the treatment of deposits owned by nonresident
foreigners, and the other the impact of recurrent Treasury financing operations on
privately owned deposits. There is, furthermore, the technical problem of properly
adjusting for collection float.
For most countries, foreignowned deposits present no major obstacle to determining the money supply.' The
situation is different in the United States, since the dollar has become a reserve
currency as well as the cornerstone of international monetary arrangements. Furthermore, a large part of world trade is financed and settled in dollars. The amount
of dollars held by foreigners for purposes unrelated to the United States domestic
economy has grown over time, especially after World War II. Yet, there is no
statistical basis for eliminating such deposits from the money supply series.
Indeed, coverage of foreign deposits in the money supply series has been widened
in recent years: first, in 1960, to include deposits owned by foreign commercial banks and subsequently, in 1962, to include those of central banks and other
foreign official institutions held by the Federal Reserve Banks. The reason given
for the increased coverage was that deposits of foreign banks and those of foreign
governments and official institutions may be used for investment or other expenditures in much the same way as balances of other holders.'
A part of the foreign deposits is related to the international role of the dollar
or to foreign operations of United States corporations. This raises a number of
questions, particularly because of the growth of world trade and of that part of
trade which is invoiced and financed in dollars, the increased role of foreign
operations of United States corporations, and the growth of the Euro-dollar
market. Even though transactions in foreign-owned dollars that are not directly

FOREIGN-OWNED DOLLARS NOT AFFECTING GNP DIRECTLY.

'In some countries of Western Europe, however, foreign-owned deposits are both large and subject
to substantial short-run fluctuations. In Germany and Switzerland,for instance, they have been subject
in recent years to higher reserve requirements and prohibitions on earning interest. Monetary statistics of most such countries, however, show foreign deposits separately.
"See Federal Reserve Bulletin (October 1960), page 1103, and (August 1962), page 944. The treatment of foreign deposits in the money supply series based on daily averages differs from that in the
last-Wednesday-of-the-month series. For a comparison, see ibid., page 945.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

101

related to domestic GNP may have been relatively small in the first years after
World War II, they are now significant enough to affect both money supply estimates(and thus Vy) as well as deposit activity (and thus Vi),as explained below."
A considerable amount of foreign balances, including those of foreign commercial banks and corporations, is connected with the trade between the United
States and other countries and thus, at least in part, to their demand for our domestic output. But other balances are held because the dollar is the leading trade and
vehicle currency, and much of the trade between third countries is billed and
financed in dollars." The level of such balances is related to the volume of trade
among foreign countries rather than to the foreign trade of the United States. Similarly, the bulk of dollar balances of foreign insurance companies and nonbank
financial institutions, of trading companies, and of foreign corporations of international scope is not associated in any significant way with the volume of demand
for our domestic output of goods and services.
Available statistical data on foreign-held balances are not complete. A bank
may not know that certain accounts are foreign, because holders frequently give a
United States address for convenience or for the purpose of concealing their
identity." Also, only deposits owned by "nonresident" holders are reported, a term
which does not cover certain resident foreigners who use dollar balances largely,
or even almost exclusively, to conduct business abroad. For example,some foreign
businesses maintain in the United States a base of operations for the entire Western
Hemisphere (or some other area larger than the United States), and others transact
foreign business, in particular when billings are in dollars, through United States
subsidiaries. Their balances do not fall into the category of "nonresident accounts".
To go a step further, most large United States corporations operate abroad, and
frequently on a worldwide basis. Their domestic balances reflect to a certain extent

"Transactions in Euro-dollars result in a large volume of reported debits since, in contrast to
domestic banks, deposits of foreign banks are included with private deposits and thus affect Vt. Furthermore,a large part of Euro-dollars borrowed by United States banks through their overseas branches
is transferred on a day-by-day basis, even when funds are retained for longer periods.
"Deposit accounts are also held for convenience, safety, and other reasons by nonresident workers,
such as Canadians and Mexicans who cross daily into the United States to live or who come periodically for seasonal employment. Such balances may or may not be ultimately repatriated and their
significance as a measure of demand for current United States output is at least uncertain. No comprehensive data on such balances are available or, in fact, obtainable.
12See Review Committee for Balance of Payments Statistics, The Balance of Payments Statistics of
the United States (Washington, D.C., 1965), page 78.

102

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

their foreign transactions and include foreign exchange operations and other
types of transfer of funds." Given the size of these activities, it would be surprising
if their dollar balances and the transactions passing through these balances did
not constitute an element of some significance in money supply and debits
statistics.
The official money supply series is confined to the private sector, because the level of United States Treasury
deposits at any given time cannot be regarded as determining the level of Federal
Government purchases and other payments, according to generally accepted reasoning. But the same can be said, in varying degrees, of other sectors of the economy as well.
Broadly speaking, there is an inverse relationship between Treasury deposits
and the private money supply. Fluctuations in tax collections and in Government
expenditures for goods and services (including transfer payments) tend to follow
regular patterns which are allowed for in seasonal adjustment factors. This is not
necessarily true for Treasury financing operations and other large, unusual, and
irregular receipts or payments. Moreover, the size of increases in Treasury deposits
resulting from financings depends on the amount of initial underwriting of new
issues by commercial banks and on the time profile of the sale of securities so
acquired to nonbank holders. Also, when the Treasury pays off debt held by
commercial banks, the subsequent effect on the money supply depends on basic
bank reserve positions and loan demand. Other recipients of payments resulting
from debt retirement may use the proceeds to reduce their indebtedness to the
banks or to buy bank-held securities. Rate levels and money market conditions
thus influence the degree to which Treasury payments tend to restore the level of
private deposits. In spite of seasonal adjustments, temporary shifts of private
deposits into United States Treasury balances (and reverse flows) may and do
produce short-run swings in private demand deposits (in some instances, due to
the incidence of payment dates for new Government securities), which are
reflected in velocity but have little or no economic significance.
ELIMINATION OF UNITED STATES TREASURY BALANCES.

"The opposite is also true: some transactions involving domestic operations, such as purchases of
imported raw materials, may be charged to dollar balances of foreign subsidiaries.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

103

concerning the adjustments for clearings and collecdistance
and delays in processing and are technical rather
result
of
float
are
the
tion
than conceptual. It is, however, not certain that the present estimating techniques
— which consist of deducting two separate items from gross demand deposits,
namely,"cash items in process of collection" and Federal Reserve float — provide
an adequate approximation of the deposit component of the money supply.
The adjustment for float, intended to eliminate the double counting of deposits,
results in only a partial correction because the reported figures used do not measure the deductible items precisely." On the one hand, reported (bank and Federal
Reserve) float is too broad, because it includes United States Treasury and interbank checks and other items not properly chargeable to accounts included in the
money supply series. Subtracting float from gross deposits produces an overadjustment by an unknown amount." It should be noted that (particularly when
banks send checks to correspondents rather than to Federal Reserve Banks) some
part of the collection float is shown in reports filed by commercial banks as "due
from banks" rather than as "cash items in process of collection"." Since, however,
the item actually deducted from gross demand deposits is "due to banks" rather
than "due from banks", the money supply tends to be overstated by the amount of
float reported in the latter item.
The various shortcomings of the deposit component of the money supply are
significant for short-run as well as for long-run comparisons. Being largely of an
institutional nature, some of these influences change only slowly over time. Although their aggregate effect on statistical totals is not normally subject to cyclical
or relatively short-run influences, significant discontinuities may occur as a result
of shifts in practices or conventions, and month-to-month changes may be influenced by purely technical factors, such as bulges in float.
FLOAT ADJUSTMENT. Problems

"The staff of the Board of Governors feels, however, that "the residual duplication ... does not
appear large enough, nor are changes in it great enough, to impair the usefulness of the data for most
analytical purposes". "Section 1: Banks and the Monetary System", Supplement to Banking & Monetary Statistics (Washington, D. C., 1962), page 7.
150ther sources of overadjustments arise from the so-called "remittance float" (bank remittances
for cash letters, which are part of the Federal Reserve float) and from checks collected from banks
open on Saturdays and nonnational holidays.
16According to FDIC Annual Report for 1959, page 50, there were 3,351 insured banks reporting that they had no cash items in process of collection on the June 10, 1959 call date. Nearly half
of all banks with deposits of less than $2 million reported no such items. See also Hobart Carr,
"Pricing Correspondent Banking Services", The Bankers Magazine, Vol. 150, No. 3 (Summer 1967).

104

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

For a variety of reasons, part of the balances shown as demand deposits is not,
in fact, available to the holders. An example of this includes those amounts held to
fulfill compensating-balance requirements by businesses that borrow; such
balances are significant, particularly in periods of tight money. Amounts similarly
immobilized include deposits held for legal purposes, such as funds in litigation
or escrow and at least part of the minimum balances in some categories of personal checking accounts. Each bank has on its ledgers a certain amount of these
funds, which are small in the aggregate and tend to be fairly stable over long
periods of time. Also, there is very little uniformity among banks with regard to
the availability of out-of-town collections to customers. Policies vary depending
on categories of depositors (corporate versus personal accounts, new versus established customers, etc.) and on internal accounting procedures. For some categories
of customers, especially corporate and other large customers, compensating
balances may, in effect, absorb uncollected balances. For others, mostly individual deposits, "good funds" are limited to those actually collected. For still
others, Federal Reserve or less liberal collection schedules apply. Not infrequently, small depositors are not given specific information on availability and
learn about collection delays only when a check is returned marked "uncollected funds".
CERTIFIED AND OFFICERS'CHECKS. Among

the various components of demand deposits adjusted, "certified and officers' checks" have grown most rapidly between
the end of 1957 and the end of 1968. This category includes money orders, checks
issued in payment of services purchased by the bank, interest and dividend checks
issued by corporate trust departments, letters of credit, and bills of exchange (the
latter being directly related to foreign trade).
While deposits of individuals, partnerships, and corporations (IPC, the main
component of demand deposits) fluctuate moderately from week to week, the
amount of certified and officers' checks outstanding is subject to fairly sharp
week-to-week changes." Fluctuations in certified and officers' checks are presum-

17From August 1967 to July 1968, for instance, such changes at New York City weekly reporting
banks averaged 14 percent, exceeding 20 percent in one or another direction in one of four weeks.
An extreme change during this period occurred in the weeks preceding and following April 10, 1968,
when the amount outstanding reached the highest level ($6,237 million), 57.2 percent and 94.4 percent higher, respectively, than on the preceding and following Wednesday. These fluctuations affect
primarily weekly and end-of-the-month rather than monthly average figures.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

105

ably in part offset by corresponding fluctuations in IPC deposits.
It is not clear which among the various components of the catch-all item "certified and officers' checks" account for the sharp increase since World War II and
for the sharp week-to-week fluctuations. Indirect evidence suggests that a large
part may be due to transactions in Euro-dollars, foreign exchange, and domestic
securities, rather than transactions related to payments for goods and services.
This is reflected by the much greater expansion in this item in New York City. IPC
deposits at large member banks in New York City advanced from the end of 1946
to the end of 1968 by 60 percent, while the amount of certified and officers' checks
grew by 412 percent. Outside New York City, the corresponding increases
amounted to 120 and 198 percent. Between December 1957 and December 1967
alone, certified and officers' checks increased in New York City by about three
times and outside New York City by only about 56 percent.
The post-World War II expansion in certified and officers' checks in New York
as well as elsewhere was precipitated by two distinct developments. One is the
growth of travel and foreign trade, which caused the issuance of traveler's checks
and letters of credit to expand. Similarly, the larger volume of capital flotations
necessitated issuance of a roughly correspondingly larger volume of certified or
officers' checks, as practically all underwriting syndicate negotiations require these
instruments. The same is true for settlement of balances in stock clearings and in
certain over-the-counter transactions. These various items may be thought of as
activized balances—i.e., means of payment "on the wing"—and are properly included with the money supply. Their volume will, of course, rise with the secularly
expanding levels of trade, travel, and financial activities to which they relate. If
between the ends of 1957 and 1968 they had increased at the same rate as IPC
deposits, by the end of 1968 their level at member banks would have been about
$5.1 billion rather than $8.6 billion.
It is likely that the relatively larger growth of certified and officers' checks also
reflects another type of bank activity, which is related to managing the banks'
reserve position rather than accommodating customers with checking accounts and
replenishing loans. When a bank obtains Euro-dollars, the lender usually asks
his bank in the United States to issue an officers' check to the borrowing bank.
Repayments of Euro-dollar borrowings normally involve issuance of "due bills"
which are not included in deposits. These items pass, however, through the clearing and collection process and show up in the banking statistics as cash items in
process of collection (float) or "due from banks". The recent growth of Eurodollar borrowings and of foreign exchange transactions had the fortuitous effect
106

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

of reducing the level of demand deposits by unknown amounts, as shown by
indirect evidence. For instance, in spite of improvements in check processing and
transportation, float at New York City member banks was in 1968 more than 2/
1
2
times as large as in 1957, while gross demand deposits rose only 69 percent. At
member banks outside New York City, float also rose more than gross demand
deposits (126 percent, compared with 51 percent), but clearly in these banks the
impact of Euro-dollar operations on float was, as one would expect, smaller. Inclusion of due bills in deductible float (and in the item "due from banks") is thus another example of the shortcomings of the money supply series, as it is impossible to
sort out those collection items which are not properly chargeable to "private demand deposits adjusted", such as Treasury checks and drafts on interbank accounts. The fact that day-to-day (or week-to-week) fluctuations in due bills may
at times be relatively large casts additional doubt on the analytical validity of very
short-run (weekly or monthly) changes in the money supply series.
STRUCTURE OF THE MONEY SUPPLY. Holdings of money
balances by the main economic sectors, such as estimates prepared quarterly in connection with flow-offunds accounts, are rough approximations, given the inadequacy of the underlying data; data are entirely lacking for deposits that are not related to economi
c
activity in the United States. Furthermore, it should also be clear from the discussion in Chapter 2 that it is neither conceptually nor statistically possible to split
private demand deposits (and, even less, money balances of any holder group)
into the "purpose" categories which have been suggested in the literature. Most
balances normally perform several functions and are neither held nor administered
with one specific or single purpose in mind.
For purely expository purposes, however, we shall accept the distinction between "transactions" balances, which are maintained at levels adequate to meet
the expected flow of payments in the near term, and all "other" (including liquidity) balances. Transactions balances are most directly related to the current level
of GNP; in recent years, they have no doubt accounted for the bulk of demand
deposits.
"Other" balances held by consumers include amounts accumulated pending
investment (or transfer to savings and other similar accounts), proceeds of investments liquidated, and balances held to meet scheduled payments at specific distant
dates (such as taxes, insurance premiums, etc.) as well as for emergencies. Similarly, "other" business balances include funds being accumulated to meet
tax,
interest, and dividend payments as well as disbursements in connection with invest-

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

107

ment projects" (including funds raised in capital markets). Municipal governments may also hold funds in excess of immediate transactions requirements."
Foreign deposits may also be regarded as consisting of two parts: those held to
support trade, production, distribution, and investment activities in the United
States, and those held for other purposes.
At least two categories of accounts are characterized by high activity: (1)
working balances of nonbank financial institutions (savings banks, savings and
loan associations, and credit unions) issuing claims that holders have to convert
into money (check or currency) before making payments, and (2) balances of
brokers and dealers in Government and other categories of securities, as well as
of investment bankers, whose GNP-connected payments are negligible in relation
to payments for securities newly issued or traded. The velocity of dealer and broker
balances is exceptionally high," in part because end-of-day balances do not reflect
the amount of additional "day" money borrowed to support such payments as
those arising from clearing arrangements for securities syndicate operations (including those which may arise from underwriting) which during a day are typically
a multiple of the closing balances.

WHAT DEPOSIT BALANCES?

Money supply statistics are based on bank ledger records adjusted to remove—as
much as possible—double counting arising from the time interval required to collect
a check after it has been deposited. It is sometimes suggested that "holder record"
totals are a more appropriate basis for analysis. These data make further allowance for the time during which checks are in the mails between payers and payees
and for various additional delays before a check is actually deposited for collection.
They are thus lower (by an amount called "mail float") than totals in the money
supply series derived from bank records.

institu"Demand balances owned by mutual funds, insurance companies, and pension and other
tional investors belong in this category.
quoted in
19In part because of inefficiency in cash management. See the study by J. R. Aronson
footnote on page 74.
"It has not proved feasible to eliminate "financial debits" front debits statistics as proposed in 1951
in George Garvy, The Development of Bank Debits and Clearings and Their Use in Economic Analysis(Washington, D.C.: Board of Governors of the Federal Reserve System,1952), page 144.

108

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

After a check is given or mailed by the payer, and before it is posted in his
account, there is usually a period when, according to bank records, the same
amount appears in two different private checking accounts; this situation occurs
especially when the check is mailed to a distant point. If checks are mailed to
payees, their subsequent collection through the banking system could, for a
variety of reasons, involve delays beyond the time normally required by the mails
and processing routines at banks. First, there is a period after the payer deducts
the check from his checkbook stub (and the amount involved thus ceases to be
part of the effective money supply) and before it is credited to another bank
account—not necessarily that of the payee—during which it is part of the mail float.
After the check has been deposited, there is a second period when the amount
appears simultaneously in the accounts of the two banks involved in the transaction
(the so-called "bank float", allowance for which is made in estimating the deposit
component of the money supply as already mentioned on page 104).
For analytical purposes, demand deposits on a holder basis may thus be structured roughly as follows:
Dh=- Da—F=Ht+H.+Bt+B.+ Mt-I-M.-FE+N+Pt ± P.
Dh = holder record balances related to changes in GNP.
D. = demand deposits adjusted.
F = mail float.
H„ B,and n = transactions balances of households, business,
and municipal governments, respectively.
H., B., and M. = "other" balances of the same three holder groups.
E = balances of investment bankers, brokers, and dealers.
N = balances of institutions issuing nonmonetary claims.
Pt = foreign balances related to transactions in the United States.
P. = other foreign balances, including foreign monetary reserves
and other official balances.
"Technological"changes in the payments mechanism (including changes in compensating balance requirements), which in the longer run are reflected in income
as well as transactions velocity, are in the main related to balances Ht, B, Mt, N,
and P. Fluctuations in the mail float (F), on the other hand, may at times cause
temporary divergencies in the movement of bank ledger and holder record
balances.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

109

It is not easy to arrive at a proper statistical adjustment for the time element in
making payments to distant points and even locally, when mails and/or checks
drawn on banks located elsewhere are involved.Since the issues are conceptual,and
not only statistical, it may be useful to review them briefly. This is, indeed, another instance in which the realities of economic processes conflict with the endeavor of analysts to describe such processes within a simple framework in an
attempt to measure the underlying magnitudes with adequate precision and to
impute specific meaning to short-term (even monthly) changes in such magnitudes.
Whether checkbook stubs (holder records) or bank ledger balances are more
pertinent in determining the behavior of various segments of the economy depends
on several factors. The most important include the frequency with which bank
statements are rendered, the proportion of out-of-town items to the total volume
of checks deposited, bank accounting procedures, and bank policies on making
out-of-town funds available to customers. The owner of a bank account can have
only an educated guess as to when the checks he issued will be charged to his
account. An individual, e.g., may gamble that his income-tax-check payment will
not clear for several weeks. But a large corporation usually receives a daily statement from its bank, and normally it will take the size and pattern of its mail float
into account in the management of its cash balance. Between these two extremes
there is a wide range of behavior and procedures. Broad generalizations on the
proper analytical treatment of the mail float are thus difficult to make since this
term covers the entire period from the issuance of a check to its being deposited
for collection, particularly as some checks in the mail float are endorsed over.
Checks which recipients intend to negotiate rather than to deposit are not recorded
in their holder record balances but are considered by their holders as the equivalent
of cash.
The practice of endorsing checks (in some cases, several times) is not limited to
households; small businessmen such as retailers, service establishments, and contractors frequently endorse checks received from customers (often checks already
once or twice endorsed) to pay suppliers. In particular, many Americans make
purchases by cashing or endorsing over Treasury checks. This is particularly true
for nearly 3 million civilian Federal employees, more than 27 million recipients of
monthly social security checks, members of the armed forces stationed within the
country, and millions of regular recipients of Federal payments of one kind or another (such as checks issued in connection with various farm programs). Checks
issued by state and lower level governmental units are also frequently held and
spent by recipients in lieu of cash. Pay checks of corporations, particularly in cities
110

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

with a few large employers, play a similar role.
The mail float is increased whenever checks are endorsed over rather than deposited immediately in the payee's account. Indeed, good arguments can be made
for including at least part of the mail float with each sector's cash balance as
measured by holder records. Circulation of endorsed checks does not affect reported (bank ledger) totals, but it reduces the volume of debits arising from a
given level of economic activity,and thus the computed velocity of demand deposits.
Both mail float and bank check float are fairly large, so that their treatment has
a significant effect on the computed series on demand deposits in the money supply,
particularly when comparisons over relatively short periods are ,involved." Bank
float alone (the two items shown in banking statistics as cash items in process of
collection and the computed Federal Reserve float) in recent years exceeded 10
percent of reported (gross) demand deposits. However, the unknown part of
checks in collection channels reported under "due from banks", rather than as
cash items in process of collection, may be even larger than the item actually
deducted. The mail float is thought to be larger than the items deducted as items
in process of collection, but there are no adequate means of measuring it." Neither
total mail float nor any of its components deemed relevant can be estimated
directly. Thus, estimates of holder balances are derived by making rigid assumptions with regard to the mail float, disregarding influences that might be unique for
any given month.
Indeed, holder record estimates (given, for instance, in flow-of-funds accounts) are derived from bank records on the basis of certain assumptions concerning the size and movement of mail float that are quite arbitrary and kept
unchanged over long periods of time." One assumption involves proportionality
in fluctuations of mail and items in process of collection, thus disregarding any

21For a detailed discussion of the two floats and the statistical problems of measuring them, see
George Garvy,"The Float in the Flow of Funds" in The Flow-of-Funds Approach to Social Accounting, Studies in Income and Wealth (Princeton, New Jersey: National Bureau of Economic Research,
1962), Volume 26, pages 431-61.
"Total check float (mail float and bank float combined) was estimated to be as large as 30 percent
of demand deposits (other than United States Government and foreign-owned balances) as per holder
records at the end of 1957. Ibid., page 439.
"For instance, consumer holder record balances, as estimated in the flow-of-funds accounts, are
assumed, on the basis of a complex reasoning, to be equal to their bank record balances. See ibid.,
page 444.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

111

specific influences that may affect mail float alone. Mail delays may affect the
former more severely than the latter, since a large proportion of items in bank
collection channels moves by air freight and on truck routes and by other means
independent of the United States postal system. Conversely, reductions in bank
float resulting from the successive shortening of Federal Reserve availability schedules do not necessarily result in simultaneous and proportionate reductions of
mailfloat.
Since only bank ledger totals can be used meaningfully for analyzing short-run
changes in the demand for money, our discussion is based on the deposit series
derived from bank rather than holder records.

Effective July 31, 1969, the Board of Governors of the Federal Reserve System
amended its regulations to require member banks to include in officers' checks and
thus gross demand deposits all transfers of Euro-dollars. As a result, beginning with
this date, the understatement of the money supply stemming from these transfers
(as explained in pages 106-7) was eliminated. It is estimated that the total amount
of these payments averaged about $3 billion per day during the month prior to
this change in the regulations; it was presumably considerably smaller at the close
of 1968 when all the series used in the present study end.

112

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Appendix II:
Share of Currency in the Money Supply
The changing share of currency and deposits in the stock of money is of some
significance in explaining variations in income velocity. Indirectly, it also affects
deposit velocity. If, for instance, a larger part of household payments is made
by
check rather than in cash, the relative weight of household payments as a determinant of deposit velocity will increase. Absence of adequate data makes it
impossible to identify the effect on velocity of shifts in the ownership composition of the
two components of the money supply, but such effects must be of some importa
nce.
Also, the changing share of currency in the money supply is, no doubt, one of
the
factors explaining divergent movements in income velocity and the transact
ions
velocity of demand deposits (see Charts 2 and 7).
The share of currency (paper bills and coin) in the private money supply has
undergone considerable change over the years. The ratio of circulating currenc
y
to total deposits had been subject to a declining trend between the end of the
Civil
War to the beginning of the Great Depression. It is estimated that at the end
of
the Civil War the public (business firms as well as consumers) held more
than $80
in currency for every $100 in total deposits at commercial banks.' By
the time the
Great Depression fell upon the country (end of 1929), only $9 in currenc
y was
held for every $100 in total deposits and $17 for every $100 of demand
deposits.
The declining importance of currency in the money supply was subject
to sharp
reversals following the banking crisis of the early thirties and during both
world
wars. Bank failures and the resultant loss of confidence in the banking
system
provide an explanation for the strong preference for currency in the
early thirties.
This caused a steep rise in the ratio of currency to demand deposits from
.18 at the
end of 1930 to more than .32 three years later. After 1933,the ratio of
currency to
money supply declined again for a variety of reasons, until World War
II brought
about a sharp increase in the relative importance of currency and the
amount of
currency outstanding arose fourfold. At the end of World War II, the
amount of

'Data for 1867 through 1945 are taken from Milton Friedman
History of the United States, 1867-1960 (Princeton, New Jersey, and Anna J. Schwartz, A Monetary
page 100, in contrast to demand deposits, currency held by the 1963). As explained in Appendix I,
United States Government is included
in the private money supply.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

113

currency in the hands of the public was excessive in relation to normal demand
grow
under peacetime conditions; it declined for a few years and did not begin to
of
amount
total
the
sixties,
the
in
ly
again until 1951! After 1951, and especial
Yet,
d.
increase
s
holding
currency in the hands of the public as well as per capita
the initial increase was so modest that the ratio of currency to demand deposits
in curdeclined and did not bottom out until the end of 1958, when $25 was held
the
than
,
however
more,
derably
s—consi
rency for every $100 in demand deposit
supply
money
the
in
y
$18 held at the end of 1930. Since then, the share of currenc
held
has been rising again,'gradually increasing so that in December 1968 $29 was
in currency for every $100 of deposits.
probably
Holdings by households (including farmers and professional persons)
and a
public,
the
of
account for about 80 percent of all currency in the hands
corpora
by
larger part of the remainder is held by unincorporated business than
trade and
tions.' A very large segment of unincorporated business consists of retail
to the
related
closely
quite
are
s
service establishments, and their currency holding
level of household expenditures.
to conSince currency is used mostly by the household sector, it can be related
d
average
public
the
of
hands
the
in
s
sumer outlay. In 1929-30 currency holding
perthe
(1945)
II
5 percent of consumer expenditures. By the end of World War
g the
centage had increased to 22 percent. Subsequent developments, includin
to decline
spreading use of charge accounts and credit cards, caused the percentage
level.
prewar
the
above
to 8 percent in 1968, which was still considerably
y-money
It is not easy to account for the successive changes in the currenc

in mid-1945, per capita cur20n a per capita basis, the story is much the same;from a level of $179 had been surpassed as per
rency holdings declined to $157 in 1960, but by mid-1968 the 1945 level
capita holdings climbed to $210.
of currency in the money supply
'This development was not unique to the United States. The share
Kingdom at the end of the war than
(defined similarly to ours) was, for instance, higher in the UnitedThis
share subsequently through 1950
before (28 percent in 1946, compared with 26 percent in 1938). steadily
thereafter, and at 33 percent
climbed
but
levels,
wartime
high
the
from
downward
d
readjuste
then the ratio has remained
in 1958 was considerably higher than toward the end of the thirties. Since
more or less unchanged.
sectors, based on a set
4AIlocations of currency by major segments of the business and household
Board of Governors for year-end
of assumptions rather than on reports or samples, were made by theued.
"Selected Liquid Asset Holddates covering the period 1934-54, but were subsequently discontin
pages 749-50. In contrast,
ings of Individuals and Business", Federal Reserve Bulletin (July 1955),
by more than 70 percent
1939
in
demand deposits held by business firms exceeded household holdings
larger), the last year
and by varying—hut much smaller—margins through 1954 (less than 20 spercent
combine currency
for which such estimates were published (ibid., page 750). Flow-of-fund estimates
with demand deposits.

114

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

supply ratio.' Nevertheless, it is possible to point to severa
l factors that appear to
have played a significant role in influencing the direction
of the ratio.
For the period between the end of the Civil War and
up to at least World War
I, the secular decline in the currency ratio presumably
bore some relation to rising
levels of personal income and urbanization. Why the
rise in income should have
caused a shift from currency payments to check payme
nts is not immediately
obvious. The gradual decline of the largely self-sufficien
t agricultural sector as a
source of income and growing urbanization are often
considered elements in
explaining the secular decline of the currency ratio.' Rising
personal incomes have
enabled households to purchase more expensive goods
(more conveniently paid
for by check). All in all, however, it seems best to regard
the income factor, along
with Cagan, as "a proxy for a host of other developments which
, on balance, may
work to increase the demand for deposits relative to curre
ncy"! Some of these
developments are not directly related to income, such as
the growth and better
performance of the banking system and the imposition of
a tax on state bank notes
in 1866 which caused these banks to encourage the use
of checks.
The growing use of service charges on small checking accoun
ts after 1933 may
explain in part why the currency ratio failed to return to the
1929-30 low.' As a
consequence of the cyclical decline in the nonwage income
component and because of New Deal reform legislation, the redistribution
of income away from
the highest quintile (the group most likely to hold large deposi
t balances) in favor
of the lower income groups may have been another factor
. Wartime dislocations,
population shifts, the sharp rise in military pay, and
income tax evasion con-

'Phillip Cagan concluded that: "These movements
cannot in the main be explained by any simple
correspondence with the trends of one or two econom
differs notably from most other monetary variables."ic factors, and in this respect the currency ratio
Total Money Supply(New York: National Bureau of See his The Demand for Currency Relative to
Determinants and Effects of Changes in the Stock of Economic Research, 1958), page 2. See also his
Money, 1875-1960(New York: National Bureau
of Economic Research, 1965) and S. L. McDonald,
"Some Factors Affecting the Increased Relativ
Use of Currency since 1939", Journal of Finance (Septe
mber 1956) as well as Solomon Shapiro,"Thee
Distribution of Deposits and Currency in the United
States", Journal of the American Statistical Association (December 1943).
6Phillip Cagan, Determinants and Effects of
s in the Stock of Money, 1875-1960, page 127.
Yet, as Cagan emphasizes, urbanization may Change
impersonal nature of urban buying discourageswork both to increase the currency ratio (since the
the use of checks and, more importantly, of credit)
and to cause it to decline (urban life provides familiar
ity with the advantages of checking accounts
and encourages the banking habit).
'Ibid., page 126.
8Too much emphasis should not be placed on
factor for, as Cagan points out, "the typical rate,
when charged, was probably well below one halfthis
of one percent", ibid., page 123.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

115

ncy during World War II. After the
tributed to the rise in the demand for curre
ditures, including the trend toward
war, the changing structure of household expen
tic) services—eating out, travel, etc.
a larger volume of personal (other than domes
machines, is likely to have con—together with the ever-widening use of vending
of currency holdings per housetributed to increasing average absolute amounts
circulation confirms some of the
hold. The rising share of coin in currency in
and relative) holdings of curexplanations frequently given for higher (absolute
inations ($500 and over)
rency, while the declining share of very large denom
the diminishing use of currency
since the end of World War II seems to confirm
for hoarding and concealment purposes.
rise in the currency ratio
Taken together, available evidence suggests that the
demand for folding money as
since World War II is not evidence of an increasing
role of money as a store of
a means of payment but is indicative of the declining
disposable income per person,
value. Measured as a multiple of weeks of average
end of the war almost conthe amount of currency has been declining since the
on to total income, average
tinuously.' The decline of currency holdings in relati
stent with the hypothesis that
weekly earnings, and consumer expenditures is consi
of currency in the total money
the main reason for the relative increase in the share
t in a shift of preferences for
supply in the recent decade or so must not be sough
transactions purposes. Rather,
holding folding rather than checkbook money for
ishing portion of redundant and
demand deposits have come to include a dimin
s, which have been shifted
reserve ("idle") funds of business firms and individual
uments and—for households
increasingly into income-earning money market instr
its.
—also into various kinds of savings and other time depos

GoverCurrency", Staff Economic Studies (Board of
'See George G. Kaufman,"The Demand for
40.
page
9,
Table
1966),
,
System
e
Reserv
l
nors of the Federa

116

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Full text of The Velocity of Money

FRASER