ABSTRACT
THE CANADIAN MONEY SUPPLY

By

Paul Elmer Polzin

The purpose of this thesis is to construct a time
series which faithfully represents the stock of money in
Canada for the 1939-1966 period. Canadian institutions
and tradition present two obstacles to obtaining a series
which corresponds to the theoretical concept of money.
First, the Canadian money producing industry consists of
both regulated and non-regulated firms producing a variety
of products, only some of which are money. Second, a
large portion of checkable deposits pay explicit interest.
Analysis of the equilibrium position of a wealth port-
folio suggests that the entire value of checkable deposits
which pay interest should not be included in the measure
of the money stock. Such deposits earn two streams of“
income, one from being used as money and the other con-
sisting of the interest payments. Only the portion of the
value of these deposits that is caused by the money stream
of income should be included in the momey stock. The re-
maining portion owe their value to the interest payments
and, like all other debt items, are not money. Empirically,
the money and the debt components may be estimated by com-

paring the rate paid on the checkable deposits to the

 

 

Paul Elmer Polzin

market interest rate. The large number of different types
of money producing institutions required a discussion of
the correct definition of reserves. It was concluded that
the only balances which should be excluded from the esti-
mate of the money stock, be counted as reserves, are those
which are used as an input for money production. Un-
fortunately, the available data was not able to separate
the money held by firms for money production from that

used as an input for non-money products, chiefly debt items.
Thus, net money production was obtained by subtracting the
entire money holdings of money producing firms from their
gross output of money. This method led to the under—
estimation of the net money output of those firms, mostly
non-chartered banks, with large non-money production.

Each type of money producer was examined individually. The
debt portion of his output was estimated and his net money
production computed. A number of quarterly and yearly
money series, both seasonally and non-seasonally adjusted,
were constructed from the output data of the money pro-
ducing institutions.

In order to test the validity of the above analysis,
demand for money specifications were run using the con—
structed money stock estimates and a similar series, used
as a comparison, which included the debt portion. The two
series yielded different regression coefficients, but, be-
cause of the similarity of the demand for money and debt, \

the results could not be used as an unambiguous test of

 

 

“I, .'»

 

 

 

 

 

Paul Elmer Polzin

the method used to construct the money stock estimates.

The only conclusion which can be held with reasonable
certainty is that the income elasticity and the rate of
adjustment of the adjusted series are greater than those of

the series which included the debt component.

 

THE CANADIAN MONEY SUPPLY

By

Paul Elmer Polzin

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Economics

1968

 

-'—v’ a... “.7

 

 

-_—'-‘ .

 

I".

ACKNOWLEDGEMENTS

A thesis is supposed to represent original work.
However, no person knows \better than the author that this
volume is simply a collection of suggestions and comments
made by colleagues, friends, and teachers. Their direct
influence is obvious in the text. Indirectly, the think-
ing of the great minds of mankind is also present. Thus,
in the interest of brevity, only those who directed the
form of this thesis, my committee, will be mentioned. To
Dr. Paul E. Smith, who nurtured my interest in quantitative
analysis, Dr. Boris Pesek, who taught me that money was
something more than a good whose supply and demand, in the
upper right-hand quadrant, determined the interest rate,
and especially Dr. Thomas Saving, who gladly undertook the
thankless job of reading my indiscernible manuscripts and
making truly constructive comments, I offer my hearfelt
thanks. Also, I would like to thank my wife, Trina, for
her insistence that I get out of the house and keep work-
ing. Otherwise, this thesis would still be in the planning

stage.

ii

 

WA‘ZUJ‘K': . w ‘

 

 

 

 

TABLE OF CONTENTS

Chapter _
I 0 INTRODUCTION 0 O O O O I O O O O O O 0

II. THE SUPPLY OF MONEY IN CANADA . . . .

Theoretical Considerations . . .

The Net Money Supply. . . . . . .
General Comments . . . .
The Total Output of Money in
The Interest Rate. . . . .

Canada.

Output of the Money Producers. . . .

The Money Supply Series . . . . .
III. SOME EMPIRICAL ISSUES. . . . . . . . .

Introduction. . . . . . . . . .

The Demand for Money in Canada.
General Comments . . . . .
The Interest Rate. . . . .
Income and Prices. . .

Adjustment Processes in the Demand

for Money. . . . . . . . .
IV 0 SUWITARY O 0 O O O O O O O O O I O O O 0

Introduction. . . . . . .

Interest Bearing and Checkable Deposits.

Construction of the Money Series.
The Demand for Money in Canada. .

APPENDIX

I. MONEY PRODUCTION IN PRE-CONFEDERATION
CANADA I O O O O O O O O O O O O I 0

Introduction . . . . . . . . . .
Commodity Money. . . . . . . . .
Paper Money. . . . . . . . . . .
II. "ACTIVE" BANK DEPOSITS. . . . . . . .
III. AVERAGE OR MARGINAL RATE? . . . . . .

SELECTED BIBLIOGRAPHY. . . . . . . . . . . .

iii

th
th
105
108

110
110
111
112
122
127

129

 

 

 

Table

11.

I2.

13.
14.
15.
16.

l7.

18.

LIST OF TABLES

The Comparison Rate. . . . . . . . . . . . .
Bank of Canada and the Royal Canadian Mint .
Chartered Banks . . . . . . . . . . . . . .
Quebec Savings Banks. . . . . . . . . . . .
Trust and Loan Companies . . . . . . . . . .
Credit Unions (Quebec) . . . . . . . . . . .
Quarterly Net Output (Qu)° . . . . . . . . .

Quarterly Net Money (Qa) . . . . . . . . . .

Quarterly Net Output, Seasonally Adjusted 02%»

Quarterly Net Money, Seasonally Adjusted (Q2).

Quarterly Net Output and Net Money,
Including Trust and Loan Companies (Qfi
and Q5) . C ‘ C C O . . C C O C . C C C .

Quarterly Net Output and Net Money,
Including Trust and Loan Companies,
Seasonally Adjusted (Qfis and Q53) . . . .

Yearly Net Output and Net Money (Mu and Ma)°

Demand for Money: Linear Specification. . .

Demand for Money: Log Specification. . . . .

Demand for Money: Percentage Change
Elasticity Estimates. . . . . . . . . . .

Demand for Money: Stock Adjustment
Specification. . . . . . . . . . . . . .

Rates of Turnover. . . . . . . . . . . . . .

iv

Page
24
28
38
#3
51
59
67
68
69
7O

71
72
73
8O
89
96

102
I26

 

 

   

 

 

 

VJ“. "v”' '

1-.

LIST OF FIGURES

Figure Page

1. Net Money and Net Output. . . . . . . . . . . . 63

2. Rate of Change of Net Money and Net Output,
Centered Moving Average. . . . . . . . . . . 6h

CHAPTER I
INTRODUCTION

Economic theory and analysis provides a set of re-
lationships between magnitudes and behavior. It purports
to predict behavior on the basis of certain predetermined
conditions or values. The purpose of empirical investigation
is to test the accuracy of the relationships and, having
confirmed their veracity, use them to makepredictions.

Rigorous testing of economic theory requires that
the empirical magnitudes closely resemble their theoretical‘
counterparts. Unfortunately, the real world does not always
provide goods with attributes similar to those employed in
theoretical analysis. Thus, the precondition for empirical
work is the existence of data which can be used to represent
the qualities employed in theoretical analysis.

Economic theory has provided lists of goods which
are critical for the explanation of different types of be—
havior. The quantity of money is a magnitude which is
predicted to have various and diverse effects throughout
the economic system. Before these causal mechanisms and
predictions can be tested, an accurate measure of the
quantity of money must be devised. The purpose of this study

is to provide a series which faithfully mirrors the

2

theoretical concept of money in Canada for the 1939-1966
period.

In the United States, almost all money is produced
by firms which are required to submit detailed reports to a
regulating authority. Except for a short period,1 the
product of these firms has closely paralleled the theoretical
concept of money. Thus, with output data readily available
in a suitable form, the construction of a series which closely
approximates the theoretical magnitude is relatively easy.

Unfortunately, there are two conditions in Canada
which make it difficult to obtain an accurate estimate of
the stock of money in existence. The first problem is a
lack of breadth in the data collected and published by the
regulatory agencies. Although all money producers are under
regulation of some sort, the regulating bodies are diverse
and independent. The independence of these bodies has
resulted in a lack of coordination in the compilation of
data. There is simply no single source which gathers and
organizes reliable information concerning all money producers
in Canada.

The lack of a single regulatory or data gathering
body is the result of actions taken by the chartered banks.

During the first half of the last century, the chartered

 

“Apparentlyg some banks in the 1920's allowed
checks to be written on time deposits. See Milton Friedman
and Anna Schwartz, A Monetary History 2; the United States
(Princeton: Princeton University Press, 1963), p. #43.

 

 

 

 

 

8Vai]

 

3

banks spent considerable time and effort in obtaining the
monopoly in the private production of notes, then the primary
form of money.2 In their zeal to procure this monopoly, they
did not attempt to restrict checkable deposits. By the 1930's,
when the chartered banks lost their monopoly in note pro-
duction and their dominant position began to ebb, there were
independent institutions with a long tradition of producing
checkable deposits. The data gathering agencies grew, and
were primarily concerned with, the chartered banks. They

had no authority, and were not disposed toward collecting
information from the non-chartered banks.

In 1939, the first year of this study, the structure
of the Canadian money producing industry had reached its
modern form. The chartered’banks were still dominant but
the "near-banks" were organized and poised for their rapid

postwar growth. The outline of the industry is as follows:

I. Th3 BE 3 9£_Canada and the ngal Canadian Mint.

The Bank of Canada and the Royal Canadian Mint are
the money producing branches of the Canadian government.
The Mint and the Bank of Canada have the monopoly in the
production of coins and notes. Both agencies publish de—
tailed information concerning their production of money.
II. I; Chartered Banks.

The chartered banks are under the regulation of the

 

2A more complete historical record of this period is
available in Appendix I.

 

A

Bank of Canada. They are required to submit detailed monthly
and weekly information regarding their Operations. In general,
the required data is complete and quite accurate.

III. Two Quebec Savings Bgnkg.

The Quebec Savings Banks Operate under a federal
charter and are subject to about the same reporting require-
ments as the chartered banks.

IV. 19 Trust and Loan Companies.

The Trust and Loan Companies are the descendants of
the Building Societies of the 19th Century. They may operate
under a federal or provincial charter. The few that have a
federal charter are required to make yearly reports to the
Superintendent of Insurance. All companies operating in
ontario must submit yearly statements to the Registrar of
Loan and Trust Corporation for the Province of Ontario. Both
agencies publish yearly summaries of the reporting companies.
Companies representing about three percent of the total of
Trust and Loan Company assets report to neither.

V. Government Savings Officgs.

During the 1920's and the 1930's the provincial
governments of Ontario and Alberta opened savings offices.
They are required to report only to their respective legis-
lative bodies and usually do not publish detailed statements.
The provincial governments have refused to release the
information needed to estimate their money production.

Consequently, they will not be included in this study.

 

5

VI. 1100 Crggit Union and Caissgg Populaires.

The Credit Union Movement began in Quebec in 1909
and has since spread throughout Canada. Most provinces have
some legislation concerning credit unions, but their data
gathering and publishing is minimal. However, many of the
credit unions belong to "centrals," which can best be de-
scribed as "central banks" for credit unions. Several of
these "centrals" provide an excellent source of reliable
data. In general, information regarding credit unions in

scarce simply because nobody collects. it.

The data of the Bank of Canada and its predecessor,
the Ministry of Finance, concerning the chartered banks
provides a fairly accurate description of the Canadian money
supply up to World War II. However, the spectacular postwar
growth of the non-chartered banks has gone unreported in
their publications.

The second difficulty with estimating the money supply
in Canada is due to the fact that Canadian money producers
have a long tradition of paying interest on certain checkable
deposits. A hypothesis will be presented in Chapter II which
suggests that, in cases such as this, the theoretical con-
cept of money is approximated only if these interest bearing
and checkable deposits are deflated by an adjustment factor.
This chapter will also present a method by which the adjust-

ment factor may be estimated.

 

6

The original, and ambitious, goal of this study was
to overcome both of the aforementioned obstacles and obtain
a true representation of the Canadian money supply. However,
as could have been predicted, difficulties arose which forced
this goal to be compromised. The raw output data of the non-
chartered banks was, after an appropriate number of setbacks,
gathered and organized. But the characteristics of the
product mix of these institutions combined with the method
employed to estimate money output to result in the under-
estimation of their net money production. Consequently, only
a shadow of their rapid postwar growth and the corresponding
increase in their influence on the Canadian economy is
mirrored in the money series constructed in Chapter II.

The plan of the study will be as follows: Chapter II
will begin with a general discussion of interest bearing and
checkable deposits and the correct procedure for their in-
clusion in the money supply. The following section will
describe, in general, the method used for the construction of
the money series. Next, the attributes of the Canadian
money producing industry will be outlined and it will be
shown how the general formulation will be adapted to the
Canadian case. The two final sections will discuss in detail
each money producer, the characteristics of the products
produced and the data available, the construction of the
money series themselves, and the secular and cyclical be-

havior of the money supply.

 

 

7

Chapter III is an empirical chapter and attempts to
set out some of the attributes of the money series constructed
in Chapter II. Difficulties were encountered, again as should
have been expected, when it was discovered that available
hypotheses were insufficient for testing the model used for
the construction of the money series. Rather than engaging
in the precise analysis required to construct apprOpriate
hypotheses, a broad sample of the results is presented with
only an intuitive explanation as to their meaning. It is
hoped that these results will provide the tests for, or the
inspiration needed for the construction of, .hypotheses at
some time in the future.

The first Appendix is the result of the author's
attempt to understand the institutional structure of the
Canadian financial system and presents the historical
beginnings of the money producing industry in Canada. The
next two appendixes discuss important tOpics which are too

short for a separate chapter but too long for a footnote.

3.5;...‘2'5 ~'. ’—-r<a.

 

._l M 4gb: 1;“4'.,

CHAPTER II

THE SUPPLY OF MONEY IN CANADA

I. Theoretical Considerations

A. General Case

The theoretical concept of money does not have an
obvious and unique real world counterpart. Economic theory
describes money as simply a good which is used to eliminate
barter and facilitate transactions. This role may be per—
formed by any one of a number of physically different
capital goods. In other words, one cannot, on the basis
of Economic theory, predict the form which money will take.

A society may set aside one good which is to be“
used exclusively as money. Or, it may use a good which can
be employed in other tasks. In the first case, the identi-
fication of money is straightforward. The second case pro-
vides some problems. If a good has a number of uses, only
one of which is money, how should the quantity of money be
determined? Should all of this good be money? None of it?
Or, should this good somehow be divided into a money and a
non-money protion?

All goods which serve as money, either partially or
wholly have in common the stream of income which eminates

from their use as money. The quantity of money, in value

8

 

9

terms, may be obtained by discounting the marginal return of
each money to the present and then summing over all goods
used as money. Let R1...Rn be the marginal return of good
q serving as money in time periods t1...tn. Then, M, the
quantity of money in value terms, is

n n
(II-l) M = 2 R. /(1 + r)

i=l all lq
q

Where r is the market rate of interest.

The money stream of income, R, is not an observable
phenomenon. If there exists a capital good which can only
be used as money, then one can conclude that its value is
equal to the discounted present value of the marginal returns.
In other words, knowing M and r, the His may be deduced.

If more than one stream of income accrues to a good
which is used as money, the entire value of this good should
not be counted as money. Assume that good X is a unit of
account and the marginal return to it as money is R1...Rn
and Pl...Pn because it has another use. The value of X
is equal to the discounted present value of the sum of the
marginal returns in each period.

(11-2) x :_
1

II M:

n n n , .n
(Ri + Pi)/(l+r) = Z Ri/(l +— r)1+2 Pi/(l+r) .
l i= 1 i=1

Only Ri is due to X being money. The discounted present

value of this stream alone measures the quantity of money.

 

r‘r...

- «1"— Tvr": V

«‘1.

 

 

10

A direct estimate of the R13 is usually impossible.
However, if the P15 are an observable market stream of income,
then the quantity of money can then be obtained by subtract-
ing the value of the P13 from the value of X,

n n n
(II-3) M 3 X - 2 Pi/(1 +r)n : z Ri/( l upr) .
i-l i=1
In the following sections, specific cases will be examined

so that an estimable equivalent to (II-3) can be obtained.

B. Case I: Pure Money
Assumptions:
(a) Reasonable certainty.
(b) Bonds (B), whose only return is explicit interest pay-

ments. Rb is the marginal payment.

V

(c Money (M), whose only return is imputed payments because
the good is money. Rm is the marginal imputed payment.

The wealth portfolio consists of only money and bonds.
This portfolio will be in equilibrium when the marginal rate
of return on money is equal to the marginal rate of return on
bonds.
(II-1+)Rm/M = Rb/B.

The entire value of M is due to its being money and

should be included in the quantity of money.

n
(II-5) M = E Rmi/(l +r)n.
i=1

 

 

11

C. Case II. Money which earns an explicit return.
Assumptions:
(a) Reasonable certainty.
(b) Bonds (B), whose only return is explicit interest pay-
ments. Rb is the marginal payment.
(0) Money (MD), which receives an imputed return from its
use as money and explicit interest payments. Rm and
Rd are the marginal imputed and explicit payments.
Equilibrium in a wealth portfolio is obtained when

the sum of the two marginal rates of return on money-debt,

MD, is equal to the marginal rate paid on bonds.

(II-6) (Rm + Rd)/ MD = Rb/B.

The value of money-debt is equal to the discounted
present value of the sum of its two streams of income.
n n n. n n
(II-7) MD: 2 (Rmi+Rbi)/(1+T): )3 Rmi/(l+r) +2 Rbi/(l+r)n:NH-D
i=1 i=1 i=1
Where M is the present value of the imputed returns, the
portion of the value of money-debt that is due to the money
services. D is the present value of the explicit interest
payments, the portion of the value of money-debt that is
not money.
The pr0portion of the value of money-debt that is due
to it being an interest bearing debt is the ratio of the
explicit interest payments to the sum of the explicit and

imputed returns.

12

 

 

n
X
(II-8) D = D = i=1 Edi/(1+?)n =ani .. Rd
D + 2 (Rmi'ARdiVfi-‘A‘fln “Ifimi+Rdi) Rm'fRd
i=1

The discounting cancels because, with the portfolio in
equilibrium, the Rdi and Rmi do not change from period to
period. Rearranging (II-6) we obtain

(11.0) Rm + Rd =(Rb/B) MD.

Then, substituting into (II-8;,
R _ R MD
(II-8') .___2____ = d - (3/
MD (Rb/B) MD Rb/B

 

The prOportion of the value of money—debt that is due to it
being a debt is simply the ratio of the rate paid on money-
debt to the rate paid on pure debt. The proportion of the

value of money-debt that should be included in the quantity

of money is

 

Rd /MD
(II-9) M = (1 - ).
MD Rb7B
The quantity of money is
Rd /MD
(II—lO) M Z ( i ) MD.

 

Rb/B

 

 

 

13

II. The Net Money Supply

 

A. General Comments

The net money supply is total money production minus
the amount of money which is used for its own production.
There is no rationale for the exclusion of money held by a
firm if it is not an input for money production. These
balances may be crucial for the determination of the
equilibrium quantity of money produced, but they should
not be excluded from the net money supply.

Chartered banks are required by law and the repurchase
clause to hold amounts of government money. Likewise, the
non-chartered banks hold government money and chartered bank
deposits. These "reserves" are a necessary input for the
production of money and their inclusion in the supply of money
available in the economy would be redundant. The general
form of the money supply estimates to be derived later is
(II-ll) M

n : Mt ' Mr‘

Mn is the net supply of money, Mt is the total production of
money, and Mr is that amount of money used as an input to
produce money.

Conspicuously absent from the net money definition is
that statement that Mn is the supply of money in the hands of
the non-bank public and that Mr is the amount of money held by
the banking system. These statements would be equivalent to

the net money definition in the case where the only product

of the money producers was money. The usual case is that

 

 

 

 

1h

money producers have a variety of products, only some of

which are money.

For simplicity, let us divide the money producer into
two parts, the first produces only money and the second
produces non-money goods. Clearly, the money held by the
money producing branch does not belong in the net money
supply. It is an input in the production of money. The non-
money producing branch may also hold balances, money may also
be an input in the production of a non-money good. However,
the rationale for the exclusion of the holdings of the money
producing branch does not apply to the other branch. Spe-
cifically, the "reserves" held by the non—money producing
branch are not used to produce money. It is only by accident
that these different operations of a money producer, producing
both money and non-money, take place under the same roof.
They are distinctly different and, for analytic purposes,
their holdings of money should not be grouped together.

Let us examine a non—monetary example. Suppose that
an electric generating company also distributes natural gas.
The net amount of electricity produced by the company, and
available for use in the economy, is to be estimated. Some
of the electric output of this company is used in generating
electricity and some is used to run the pumps on the gas
pipeline. The electricity used up in the generation plant
clearly should not be included in net production, it is an

input used to produce electricity. However, the power which

15
is used to operate the pumps belongs in net production,
available for general use. It is only by accident that the
natural gas Operation is run by the same company that
produces electricity.

Let us take a bank which has $200 of non-checkable
deposits and $10 of demand deposits. It prefers to hold
"reserves" of 20% against the demand deposits and 5% against
the non—checkable deposits. If all the money holdings of
this bank were excluded from the money supply, the absurd
conclusion would be reached that this bank has a net money
output of - 2 dollars.

Failure to include the "reserves" held against non-
money production in the net money supply will result in
absolute and relative underestimation of the money manu-
factured by individual and groups of producers. Let Mm be
the amount of money; used as an input for the production of
a non-money good. Then (II-ll) becomes
(II—ll') Mn = Mt - Mr + Mm.

Mr + Mm is the total money holdings of a money producer. If
Mm is erroneously assumed to be zero, subtracting all the
"reserves" of a money producer from total money output will
underestimate net money production.

Let us assume two money producers. The first pro-
duces only money and the second produces money and a non-
money product which uses money as an input. (II-l2) repre-
sents the net money production of the first producer and

(II-13) that of the second.

  

 
 

_- w.’- -—p-- - I

 

16

(11-13) an Mt2 - Mr2 + Mmz. Mm2 > 0-
If Mm2 is erroneously assumed to be zero, and all the money
holdings of the second producer is subtracted, the total net
production of money, Mnl + an, and the relative net money
production of the second producer, Mug/Mnl: are underestimated.
Let us assume that both producers have a non-money
product and the amount of "reserves" held against it is
directly related to output. If both producers have the same
money output but the second has a larger non-money output,
then the relative net money production of this producer is
underestimated. Thus, the degree of underestimation depends
on the ratio of money production to the total output of
all products of the producers.1
The non-money producing branch of a bank is
analytically distinct from the money producing branch. If

the "reserves" of the non-money branch are to be excluded

from the money supply, logic would demand that the money

 

1Consequently, the net money production of the various
types of producers which will be derived later cannot be
legitimately added to the growing body of Canadian literature
concerned with the extension of government control to the non-
chartered banks. One form which this argument has taken seems
to imply that when the market share of the non-chartered banks,
where the market is variously defined as the money market or
the market for all financial assets, reaches some critical
level, government regulation should be extended. The above
analysis implies that, given the different proportions of the
money - non-money product mix of various producers, the degree
to which the true money production is underestimated will
vary between producers. Thus, the net money series of this

 

 

l7

holdings Of other firms which produce similar products
should also be excluded. Then we would have the "money sup-
ply outside banks, sales finance companies, insurance cor-

porations, etc."

B. The Net Money Supply in Canada.

During the period under study, the media of exchange
in Canada were produced both domestically and abroad. Al-
though the use of foreign media may be considerable, such as
traveler's checks and American currency, the complete lack
of information concerning their magnitudes forces their ex-
clusion from this study.

The Canadian money industry produces (1) three types
of money, coin, notes, and non-interest bearing and checkable
deposits, (2) one non-money product, debt instruments, and
(3) one joint products, money-debt. This section will at-
tempt to show how the outputs of these producers can be
combined to represent the net money supply in Canada.

1. The Total Output of Money in Canada.

Coin, notes, and non-interest bearing deposits will
be assumed to be pure money. Specifically, it is assumed
that the only stream of income eminating from these goods is

that which results from their use as money. The implication

 

study will not provide unbiased estimates of the market shares
of the different producers. See Canada, Royal Commission on
Banking and Finance, Report (Ottowa: Queen's Printer, 1965),
Chapter 6. and J. F. Graham, et. al., The Role of the Trust
and Loan Companies in the Canadian— Economy (London: School of
Business Administration, University of Western Ontario, 1965),
Part II.

 

 

 

18

of this assumption is that the positive and negative income
streams produced by the services and costs of demand deposits
are zero. This assumption being made, the entire value of
these goods will be included in the output of money. Let C

be coin, N be notes produced by the government, Nb be notes

g
produced by private producers, and DD be checkable and non-
interest bearing deposits. Then, this portion of money out-
put can be represented by

(11—11.) c ., N + Nb .1. DD.

g

The money-debt produced in Canada consists of interest
bearing demand deposits, or, a more accurate statement might
be, time deposits transferable by cheque. Due to the actions
of the money-debt producers, the holdings of money-debt pro-
ducers release statements only quarterly or semi—annually, the
marginal costs of checks rises very quickly, and they attempt
to convince business or high turnover accounts to use Current
Accounts (pure demand deposits). Thus, the owners of money-
debt are mostly individuals. The amount held by business or
commercial establishments is minimal.2

Let MD be the output of money-debt, rmd the explicit
333$ paid on money—debt, and r a weighted average of rates
paid on pure debt items. Then, the contribution of money—
debt to the total money output is
(II-l5) (l - rmd/r) MD.

The final type of product produced by the Canadian

 

2For complete discussion of the attributes of
Canadian money-debt, see below.

 

”mtg-v" " ' -

 

 

19

money producers are pure debt instruments. They are not media

of exchange and will not be included in the total money pro-
duction in Canada.

Thus, the total output Of money in Canada may be
represented by
(11-16) c + Ng + Nb + DD + (l-rmd/r) MD.

2. Mongy Used as an Input.

The need to include the "reserves" held against non—
money production in the net money supply has been discussed.
Unfortunately, all attempts to estimate the money input of
non—money production has failed.

The required reserves, of those money producers have
reserve requirements, are a percentage of total output.3
Thus, there are no required reserve ratios for money and non-
money production.

Numerous attempts were made to estimate the "reserves"
held against money and non—money output for various types of
producers. The results of many different formulations of
cross sectional analysis were either insignificant or un—
reasonable on a priori grounds.

Another estimation procedure would be to investigate
several individual producers which sell either money or a
non-money exclusively. However, generalizing on the basis of
these few examples would be very risky. One would not only

have to generalize within types of money producers, but

 

3For a historical discussion of this practice see
below, Appendix I.

 

_.
.m:
_w
. “'1!— T
‘79-“
.4: '

, -. all?“ “

 

.- IIIII .—-ru-u.‘-—F ~-#_-I

 

20

between very different classes of institutions. Certainly
the desired reserve ratios of a credit union in Togo,
Saskatchewan are different from the Bank of Montreal. The
insignificant results may also be explained by the lack of

a consistent relation of "reserves" to output between pro-
ducers. One researcher concluded that for the Trust and
Loan Companies "there is no common pattern in these percent-
age figures [desired reserve ratios] reported to us."[+ If

a stable function within a classification is unobtainable,
generalizing between types of producers is not justifiable.

The net money supply will be estimated by subtracting
the total money balances of the producers from money proa
duction. The assumption being: the "reserves" held against
non-money production are zero. This supposition is probably
erroneous. Thus, the aforementioned warnings concerning the
underestimations resulting from this method should be kept
in mind.

3. The Net Mongy Supply in Canada.

In the two previous sections, the total output of
money and the amount of money used as an input for money pro-
duction were derived. Combining them we have the general
formulation of the net money supply in Canada,

(11-17) C + Ng + Nb + DD+(l — rmd/r) MD um.

Where TR is the total money holdings of money producers.

 

4J. F. Graham, §E_§l., Role, p. 11-36.

_—._

 

 

21

C. The Interest Rate

The first sections of this chapter have outlined the
importance of the interest rate in determining the supply of
money. Unfortunately, the empirical counterpart to "the
interest rate" is not obvious. The range of rates which do
exist exhibit a spectrum of values due to differences in
risk and convenience premiums. The optimum rate, for the
purpose of analyzing money-debt, would be one which had ap-
proximately the same risk and convenience premium as
money—debt.

The money-debt produced in Canada is a fixed price
asset whose ownership is effectively restricted to indi-
viduals. The correct comparison rate would be one on a short
term fixed price pure debt instrument which is also sold
mainly to individuals. Thus, the risk and convenience premi-
ums of both goods would be approximately equal.

The best comparison rate would be that paid on the
non-checkable deposits of the Trust and Loan Companies and
several credit unions. Unfortunately, these rates are not
available in reliable form. Consequently, the compaflson
rate used in this study is a weighted average of a rather
arbitrary choice of pure debt items which seem to meet the
criteria in one way or another.

The first item in this array is the Debentures and
Guaranteed Investment Certificates of the Loan and Trust

Companies. The Loan Companies' Debentures are sold over the

 

 

 

 

22

counter with maturities over oné year and, in a few cases,
up to ten years. The average term to maturity being under
five years.5 The Guaranteed Investment Certificates of the
Trust Companies have a much shorter term to maturity, but a
lower proportion are held by individuals. Almost all mature
within five years of the time of issue, with most coming due
within one year.6 The majority of the holders Of both
Debentures and G.I.C.s are individuals.7

The other item used in computing the comparison rate
is credit union shares. The use of this rate is subject to
several serious shortcomings. The reliable data available
is restricued to the Caisses Populaires of Quebec and credit
unions is Saskatchewan. Even if it is assumed that the rates
presented are representative of rates paid by all credit
unions, they may not reflect an alternative open to all
Canadians. Credit unions are concentrated in those locali-
ties where the "credit union movement" has had a strong
emotional appeal. Also, many credit unions, in an attempt
to restrict domination by a few members, allow each member
to own only one share.8 This may result in these shares
paying a lower rate than would otherwise prevail. The mem-

bers could view the low rate as being the cost of having

 

5Royal Commission, Report, p. 179. 6Ibid. p. 182.
7J. F. Graham, et al., Role, p. 11-25.

 

8Royal Commission, Report, p. 159.

 

23

the access to the credit union facilities. However, all

shares may be withdrawn on demand or with short notice.

Canada Savings Bonds are not included in the com-
putation of the comparison rate. They are similar to Series
E bonds in the United States in that they are sold at a dis-
count and pay the stated rate if held to maturity. Infor-
mation concerning their stated rate is available. However,
there is no data concerning the effective rate or the
average time a bond is held until retirement.

The comparison rates presented in Tabel II-l are
weighted averages of the annual rates paid on Debentures,
G.I.C.s, and credit union shares with the weights being the
annual average of the item outstanding. It may be argued
that a different weighting scheme may be in order. For
example, one may want to weight shares more than Trust and
Loan Company issues because of the latter's long term to
maturity. However, after all factors were considered, it
was felt that the weighting pattern used provides a good
index of rates available to individuals at a minimum cost

of time and effort.

D. Output of the Money Producers
(l. Governpppt Production
3. Bank of Canada
The primary government producer of money is the
Bank of Canada. It was incorporated under the Bank of

Canada Act of 193A and began operations in March of

 

 

 

 

 

mm.¢ Hm.: m.mm sa.m ©.mm nw.q m.mm© ms.s o.H©m Hwoa
ms.s ss.s m.os no.m s.sm mp.s m.cpm Hq.s o.mms QOOH
so.s mm.s o.ma m©.m m.ms sm.s m.oss mm.m m.mmm mmoa
mo.m 0H.s w.nm os.m 0.0m ow.s m.ums ms.m m.mds mmma
do.m 00.: mosm do.m 6.0m mm.m m.w©m H©.m m.mom smdfi
as.m om.m 6.0m mo.m m.mm as.m m.6sm Ha.m m.©mm @de
ma.m Ho.m m.sm oo.m m.mm Hp.m m.wom sn.m m.mmm mmma
om.m mu.m m.sm m©.m m.om sp.m m.wpm s©.m m.m©m smoa
mo.m ms.m o.am op.m 0.6H cm.m m.wam on.m o.©om mmda
mm.m mn.m m.©a ma.m o.ma ma.m m.wma as.m w.mma mmoa
os.m mm.m s.ma os.m H.o ©©.N m.woa Hm.m 0.05H Hmoa
as.m s©.m 0.0H mp.m 0.8 mm.m m.msa so.a o.osa Omma
mm.m oo.m H.aa m©.m m.© mm.m m.mma ©H.N o.HmH dead
m©.m mm.m s.ma mm.m s.s md.m 0.4HH pm.m m.mma msoa
nn.w Om.m 0.0H mm.m m.m mo.m o.moa ms.m m.ama mqoa
mm sm.m mm.m m.m m©.m ©.m mm.m o.q© 5:.m o.moa oada
Ho.m sm.m s.© H©.m s.H mm.m 0.0m N©.m m.aa mama
mo.m mm.m m.m om.m a. ms.m o.Hd as.m m.©m ssma
mH.m Ho.m m.m mm.m m. ms.m m.mm os.m m.mm mama
Hm.m om.m m.m oa.m m. ms.m ©.md on.m 0.05 mama
mm.m wm.s n.m mo.s H. mm.m m.w© m©.m o.mm Haoa
sm.m m4.s s.m us.m H.ooa mm.m s.om osma
m0.m ss.s N.m Ha.m s.aoa em.m m.sm amam.
3mm $23.95 3mm AmCOHHHEV 3mm Amcoflaaflsv 3mm AmmOHasv
mpmm mmsmgm mpmm mmpmsm mpmm mopSpcmme mpmm .o.H.o
d m s o m a m m H
Ompmso cmzmnopmxmmm
. mpmm somflpmasoo mcoflcs pflpmpo mmflcmmsoo smog mmflcmasoo pmzpb pmow
me<m zomHm<mEoo Mme
H mqm<e

 

 

25

.0 mm osmm

.oococcommmppoo

Hmsomhmm .oonwso .mfl>mq .mCvammmwo mmnfimasaom mommflmo moo coapmpmpmm MA

.Ampmoh pmpma was od0a .mhdeOOHnw¢ mo paws

upnmgma "mzmppov mvmcmo cw mcoflss pfipmpo .osSOHSOpr< mo psmspnmmmm .mnmcmo

.hmswwz pmpma cam 040a .moapmflpmpm mo :mohsm ompodo ”hpfio oonmsov

stcc< Hmowpmflpmpm omQOSG .mOHpmemum mo :mmpsm omnwdo .oonmzc .ooma Op Homa

Eopm mcfippmnmoo mmsflmasaom mommamo haco mopsaocH .om mush so mono mommfimo
30mm mo pmwh Hmomfim wnp pmsu pm55mmm ma pH .cofimeSQEOo mo mmoaudm can now

 

.m mm seam

.oocopsommohpoo
Hm:OMhmm .ucosQOHo>oo o>flpmpmgouoo new compmemmouoo mo pewspemgwn .cmzmsopmxmmm

.m as camm
.H mm mamm

.Amhmmh
pmpma cam mm0H .nwpsflnm m.:mmsd "Opconoev pmommm .mmwsmmsoo undue cum smog
mo panamawom .oapmuco .owpmpco CH wsfipmpmno mmflcmasoo smog paw pmSpB haso mopsaocH

.Ampmmh hwpma new

©40H .mcmcmo mo xcmmnmzmppov hamEESm HmOHpmempm .mpmcmo mo xsmm .mpmcmo

.Amhmm% pmpma cam mm0a .hopsfihm

m.smono Househoev psomwm .mwficmmsoo pmSMH cam smog MO pmhpmwwwm .OHMMpCO
.oasmpso ca waflpmpwmo moacmmsoo smog paw awash >Hco mopsaonfl .ms0au0m0a “moonsom

.m

.m
.¢
.m

 

N4.m 00.: 0.00H sm.0 0.0ma as.m .msea 04.0 0.0Hmm oo0a
No.0 00.0 0.x0 00.0 s.moa 00.0 0.0soa 00.: m.HNOH mo0H
mm.s so.0 s..0 00.0 H.00H 00.0 0.600 s0.s 0.00ea 400a
40.: Om.m 0.00 00.0 0.0HH 00.0 0.0Hm mm.s m.mmma mp0H
00.4 OH.0 m.so 00.0 0.00 40.0 0.0H0 00.0 o.0aoa No0a

.qr‘

 

.st-rn

A.»- w s’i‘rrfls‘t'.

 

 

 

26
1938.9 It is charged with the sole responsibility for regu-

lating "credit and currency in the best interest of the
economic life of the nation.10 In order to fulfill this
obligation it was given the apprOpriate powers concerning the
printing of notes and the regulation of bank output.11
The two types of money produced by this agency are
Bank of Canada Notes and Deposits. Prior to the formation of
the Bank of Canada, the primary note production of the govern-

12 The 1934 Act transferred the

ment was Dominion Notes.
"liability" of these notes to the Bank of Canada and it im-
mediately began to replace them with its own issue.13 The
series used in this study is the month end reportings of the
Bank of Canada's "Notes in Circulation." In the early years,
this series includes both the Bank of Canada Notes and the
Dominion Notes not yet retired.

The Bank of Canada accepts deposits from chartered
banks, Quebec Savings Banks, the Government of Canada and
its enterprizes, foreign central banks, and the provincial

governments.lh All deposits are checkable and none bear an

 

9Canada, Dominion Bureau of Statistics, The Canada
Yearbook,_l952 (Ottowa:Her Majesty's Printer, 1958) p. 1135.

10The Canada Yearbook, 1965, p. 1030. 11Ibid.
12The Canada YearBook, 1959, p. 1108.
13Dennis deMelto, The Supply of Money in Canada

1867-1961 (unpublished M.A. thesis, McGilI University, 1961),
p. 103.

 

thanada, Bank of Canada, Statistical Summary, 1966
(Ottowa: Bank of Canada, 1967) p. In.

 

 

 

27
explicit rate of return.15 Consequently, the full value of

the deposits will be included in the output of money. The
series used in the month end report of total Canadian dollar
deposits at the Bank of Canada.

Until 19AO the Bank of Canada was required to hold
gold "reserves" of not less than 25% of its outstanding notes
and deposits. This requirement was suspended in l9h0 and
abolished in 1952.16 Gold has always been "legal tender"
in Canada, but only a few gold coins were produced and, dur-
ing the period under study, very little gold circulated as
a medium of exchange.17 Consequently, it will be assumed
that gold is not money. Thus, the gold holdings of the Bank
of Canada will not be subtracted from money output to Obtain
net money production.

b. The Royal Canadian Mint

The Royal Canadian Mint produces silver, nickel, and
bronze coins of various denominations less than one dollar.18
Unfortunately, the only output information available is the
month end reports of the "coins in circulation outside chartered
banks." The lack of information concerning the chartered
banks' holdings of coin will lead to an underestimation of the

net money production of the Mint and an overestimation,

 

15The Canada Yearbook,gl96l, p. 1108. léIbid., p.1109.

17E. L. Stewart Patterson, Canadian Banking (Toronto:
The Ryerson Press, 1940) p. 90.

 

18The Canada Yearbook4_1937. p. 888.

TABLE 2

 

BANK OF CANADA AND THE ROYAL CANADIAN MINT

 

 

End of Net Output and
Quarter Notes and Coin Deposits Net Money
1939 I 190 233 423
II 198 231 429
III 245 270 515
IV 267 281 51,8
1940 I 255 261 516
II 309 250 559
III 369 245 614
IV 398 239 637
1941 I 404 233 637
II 433 264 697
III 472 339 811
IV 538 312 850
1942 I 551 367 918
II 585 279 864
III 672 289 961
IV 742 331 1073
1943 I 768 307 1075
II 809 410 1219
III 868 353 1221
IV 930 378 1308
1944 I 953 448 1401
II 977 445 1422
III 1041 510 1551
IV 1096 443 1539
1945 I 1108 494 1602
II 1125 572 1697
III 1175 505 1680
IV 1192 704 1896
1946 I 1166 756 1922
II 1178 644 1822
III 1211 626 1837
IV 1251 720 1971
1947 I 1217 761 1978
II 1218 634 1852
III 1237 671' 1908
IV 1278 673 ‘1951
1948 I 1247 648 1895
II 1274 762 2036
III 1335 716 2051
IV 1359 726 2085
1949 I 1314 688 2002
II 1341 754 2095
III 1363 817 2180
IV 1381 700 2081

2a

 

TABLE 2--Continued

 

 

End of Net Output and
Quarter Notes and Coins Deposits Net Money
1950 I 1331 789 2120
II 1351 795 2146
III 1395 836 2231
IV 1445 811 2256
1951 I 1398 830 2228
II 1432 886 2318
III 1470 825 2295
IV 1548 780 2328
1952 I 1477 737 2214
II 1527 653 2180
III 1562 669 2231
IV 1649 687 2336
1953 I 1581 728 2309
II 1617 779 2396
III 1645 670 2315
IV 1693 706 2399
1954 I 1602 771 2373
II 1648 764 2412
III 1679 629 2308
IV 1720 617 2337
1955 I 1649 643 2291
II 1717 684 2401
III 1765 658 2423
IV 1840 674 251p
1956 I 1779 585 2364
II 1839 658 2497
III 1876 655 2531
IV 1977 582 2559
1957 I 1832 623 2455
II 1894 619 2513
III 1929 577 2506
IV 2016 584 2600
1958 I 1923 659 2582
II 1993 702 2695
III 2028 779 2807
IV 2119 723 2842
1959 I 2041 704 2745
II 2094 704 2798
III 2101 738 2839
IV 2149 718 2867
1960 I 2047 696 2743
II 2129 670 2799
III 2146 660 2806
IV 2206 731 2937

 

30
TABLE 2-—Continued

 

 

End of‘ Net Output and
Quarter Notes and Coins Deposits Net Money
1961 I 2125 723 2848
II 2168 698 2866
III 2214 793 3007
'IV 2305 823 3128
1962 I 2179 825 3004
II 2275 851 3126
III 2304 838 3142
IV 2411 827 3238
1963 I 2278 822 3100
II 2359 868 3227
III 2381 876 3257
IV 2503 899 3402
1964 I 2398 931 3329
II 2465 906 ' 3371
III 2497 958 3455
IV 2610 987 3597
1965 I 2519 960 3479
II 262'7 1024 3651
III 2671 1043 3714
IV 2798 1185 3983
1966 I 2690 1104 3794
II 2800 1087 3887
III 2843 1142 398
IV 3022 1174 4196

 

 

a. Coin outside banks.

Sources:

Canada, Bank Of Canada,

Statistical Summar (Ottawa:
Bank of Canada, 1946 and later years).

Dennis deMelto, The Supply of Money in Can
(unpublished M. A. thesis, McGill University, 19 ,

Table 1.

ads 1867-1961
3

 

 

31

because their coin holdings are not subtracted from money

output, of that of chartered banks. These errors will cancel
when the government sector is combined with the chartered
bank sector.

The assumption that neither government producer
holds "reserves" leads to both the output and the net pro-
duction of money being represented by the sum of Bank of

Canada Notes, Deposits, and the Mint's production of coin.

a. Pure Money

(1)Nppp§ - The chartered banks worked hard and long
to obtain the monopoly in the private production of notes.19
However, their position was gradually erroded by increased
government production. With the formation of the Bank of
Canada, in 1934, the license of the chartered banks to pro-
duce notes was revoked. By 1945 they were required to reduce
their issue of notes to an amount not in excess of their
paid-up capital. After January 1, 1945 they could issue or
reissue no new notes. In 1950, the Bank of Canada paid the
chartered banks a sum equal to the value of the unredeemed
notes.20 The private production of notes in Canada was
brought to an end.

For the 1939 to 1950 period, the series used is the

"total notes" of chartered banks. From 1950 to 1966, the

 

l9See below, Appendix I.
20The Canada ngrbook, 1955. p. 1191.

 

32

value of the unredeemed bank notes are included in the Bank
of Canada note production.

The practice of the chartered banks was not to re-
issue notes of other chartered banks which came into their
possession.21 These notes were sent through the clearing
houses for redemption.22 Thus, the holdings of notes of
other chartered banks would be included in Float.

(2) Public Demand Deposits - Public Demand Deposits,
or Current Accounts, at chartered banks are essentially
identical to demand deposits in this country. They are non-
interest bearing and checkable deposits.

The"service" charges on these deposits vary between
banks and over time. The usual procedure is a constant charge
per deposit or withdrawal with a minimum per month charge.

A certain number of free withdrawals or deposits are allowed
depending on minimum monthly balances. The bank also reserves
the right to assess additional "service" charges if an

individual account warrants it. Cancelled checks and a

 

211t is usually more profitable for a bank to sell
its own money than to resell the money produced by another
bank. If a bank receives a dollar of another bank's money
and then returns it to circulation, the difference between
the purchase and sale price would, in equilibrium, only be
equal to the value of the intermediation services. However,
if the receiving bank returned the money to its producer and
requested dominant money, this bank could use the receipts
as "reserves" against the sale Of a larger amount ot its own
money. The small number of chartered banks and the well
organized clearing process has provided a low cost mechanism
for the redemption procedure. For the historical beginnings,
see below, Appendix I.

22Patterson, Canadian Banking, p. 95.

 

 

 

33

statement are sent out monthly or more often if requested.23

In addition to Demand Deposits, chartered banks also
offer Personal Chequing Accounts. These are non-interest
bearing and checkable deposits available to individuals.
They offer a lower cost per check than Personal Savings
Deposits and regular statements with the return of cancelled
checks.24 They have not been very popular and amount to
only 2% of Current Accounts.25 Both Personal Chequing Ac-
counts and Current Accounts are included in the Pure Money
series.

(3) Government Deposits - In Canada, as in most

 

countries, the various governmental units hold balances in
the commercial banking system. The usual practice is to
exclude these deposits from the supply of money. The
rationale for this exclusion is usually not stated, but

the implicit reasoning is that the money held by a gov-
ernmental unit is somehow different from that "in the hands
of the public." Extending this analysis further, one would
reach the absurd conclusion that all capital goods held by
a government should not be included in the net wealth of
the nation.

The analysis of this_study has concluded that the

 

23The Canadian Banker's Association, Submissions to
the RO al Commission on Banking and Finance (Toronto:Canadian
Banker's Association, I9637,pp. 44-45. ,

241bid. 25Royal Commission, Report, p. 118.

 

34

only money which should be excluded from the net money sup-

ply is that which is held as an input for money production.
The deposits held by the Dominion Government are used for
transactions, as is the money held by the public, and do
not seem to be viewed as "reserves" against Bank of Canada
or Royal Canadian Mint production of money. Likewise, the
deposits of Provincial Governments cannot be easily classi-
fied as a money input. Consequently, government deposits
will be treated in the same manner as any other money pro-
duced in Canada and included in the net money supply.26
(1) Government of Canada - The deposits of the

Dominion Government are allocated among the chartered banks
with balances over an agreed minimum paying an explicit
interest. An extensive search has been able to uncover only
a meager outline of the rates paid on these deposits. In
1959 "they paid 10% below the average rate paid on accepted
tenders at the weekly auction of Treasury Bills ... on the
amount by which the Government's weekly balance at all banks
exceeds $100 millions."27

For this study, it will be assumed that no interest
is paid on these deposits. Thus, they are pure money and

included in the output of money at full value.

 

26For a similar treatment of the American case, see

Boris P. Pesek and Thomas R. Saving, The Foundation of Money .

and Banking, (New York: The Macmillan Company,1968), p.223.

27Canadian Banker's Association, Submission, p. 45.

 

 

35

(ii) Provincial Government - Provincial Government
Deposits are simply demand deposits held by the provincial gov-
ernment.29 Consequently, they will be considered as pure
money and included in the output of money at full value.

(4) Deposits by Other Banks - The nationwide character
of the Canadian chartered banks enables them to conduct
domestic business without correspondent banks.29 Also, all
clearing house deficiencies are paid with Bank of Canada
balances. The deposits reported in this classification are
the Canadian dollar deposits of foreign banks.30

The foreign ownership of these deposits, while they
may have an important bearing on other economic problems, do
not qualify them for exclusion from the net money supply.
Specifically, they are not used as an input to produce money.

As best as can be determined, no interest is paid
on these deposits. Therefore, no adjustment will be made
before they are included in the output of money.

Unfortunately, this series is not available before
1946.

b. Money Debt: Personal Savings Deposits.
The nominal rate is uniform and changes simultaneously

throughout the country because it is set by the Canadian

 

2"’Ibid” p. 46.

290. E. Freeman, Chief, Research Department, Bank of
Canada, Personal letter, February 28, 1968.

3OIbid.

 

36

Banker's Association.31 The interest payments are determined
by applying the nominal rate to the minimum quarterly balances
and are paid semi-annually.32

The policy of the chartered banks has been to use
direct and indirect means to restrict the ownership of
Personal Savings Deposits to individuals. The depositor
does not receive a statement nor cancelled checks. He is
presently allowed one free check per $100 of minimum quarterly
balance and 15 cents is charged for each additional check.33
Apparently, the banks also have the right to convert Personal
Savings Deposits into Current Accounts if they become "too
active." More than 3 or 4 checks a month was azone time
thought sufficient for an account to be examined.34

The attempt of the chartered banks to control the
ownership of Personal Savings Deposits has been quite suc-
cessful. In 1961, the average size of an account was $711
and 84% were below 1,000.35 It is for these reasons that
Personal Savings Deposits are considered consumer wealth
items with an explicit rate reflecting the risk and conveni-

ence premiums associated with other items primarily intended

 

31Joseph French Johnson, The Canadian Banking Systgm,
(Washington: Government Printing Office, 19107, p. 134.

320anadian Bankers Association, Submission, p. 48.

 

33Ibid., p. 44—48. 33Patterson, Canadian Baking, p.122.

 

35Canadian Bankers Association, Submission, p. 48.

 

37
36

for individual's portfolios.
0. Pure Debt: Other Notice Deposits.

Other Notice Deposits is a catchall classification
for a number of debt items sold by the chartered banks.
Prior to 1939, the chartered banks were required to classify
private deposits only as Demand or Notice Deposits. Begin-
ning in that year, Notice Deposits were further divided into
Personal Savings and Other Notice Deposits. A result of this
decision was to segregate the checkable from the non-
checkable deposits. However, the division was not complete
and the first item mentioned below probably includes some
checkable deposits. This study will abide with the spirit
of the Bank of Canada classification and assume that Other
Notice Deposits are all non-checkable. Therefore, they
are not included in the output of money.

The first type of deposit in this classification
is "the savings deposits of institutions and corporations
which are maintained, charged service charges, and paid
interest in exactly the same way as Personal Savings
Deposits."37 These deposits are probably used by small

business as a short term and highly liquid asset.

 

36Because it is assumed that the rate paid on money-
debt reflects the same convenience and risk premiums as other
consumer debt items, the rates paid on these issues, rather
than a "true" interest rate, are used to represent the
alternative cost of holding money-debt. See above.

37Canadian Banker's Association, Submission, p. 49.

 

 

 

 

 

 

 

 

HOHm omos :mma om.a mnam :m0 H0©m >H
osnm m0©m mema om.a #HNN own mnmm HHH
mm0m 0H0m 04HH omoa 000a 80m @06m HH
000m 0600 0oHH oN.H 000a mom ooam H 840a
o0om 00mm qaoa ©N.H m0©a 0N0 000m >H
004m 00am 0moa ©N.H 0m0a 0mm 000m HHH
600m Hmam 0H0 om.a mm0a 0m© mmmm HH
mama 000m 000 60.H oaoa 000 a0ma H 040a
0mmm 000m 000 0N.H mmsfi H00 0:0H >H
anam m00m ~00 0m.a 0HOH mm0 mpma HHH
NO0H Hmdm 04m 0m.a 000H 004 md0a HH
0HHN Hmom 0mm 0N.H 0sma m0: #mma H mq0fi
msma 000m 0m0 0N.H dmsa 000 mesa >H.

.mp 0H0a omem Q00 00.H Homfi was ma0a HHH
©00H 004m 000 0N.H d0mfi mms mmoa HH
0mmH 000m :00 0N.H NHOH 00¢ ©0NH H as0a
006a 00am 000 om.H H0sa was mama >H
mood 00am mm0 Om.H 0©¢H 04: mmaa HHH
000a :mam mo0 Om.H 0H¢H 0mm NOHH HH
m0oH 00mm 000 Om.H Hmsa 000 0HHH H 080a
000a 0Hmm H00 Hm.H HO0H 8m: m0oa >H
0NOH 000m H00 Hm.a 000a N0m m00 HHH
NHOH HOON s00 Hm.a m0sa 000 N00 HH
m0¢a mmom mo0 H0.H 4HOH com 400 H 0m0H

Amcowaaflzv
090D ApCmO
Amcoaaaflev AmCOHHHHEv 00:02 mo gummy pnmo mmCOflHHHEv AmCOHaHAEV AmCOHaaflev pmpgmda
_ >0:o2 pmz psapso pmz COprom pakmcoz co 66906300202 9 pmoam nSflm m mmcoz 00:0 00 6:0
00:02 60mg mumm mm>pmmvm V
mxz<m mmmmem<mo
m mqm<e

- umw‘wyv—. ‘—

 

39

0N.H
4N.H
4m.H
4N.H
#N.H
4N.H
4m.H
dw.a
©N.H
0N.H
©N.H
©N.H
0N.H
0N.a
0N.H
0N.H
0N.H
0N.H
0N.H
0N.H
0N.H
0N.H
0N.H
0N.H
am.a
¢N.H
sm.H
sm.a
HN.H
HN.H
HN.H
AN.H
MN.H
mm.H
mm.H
mm.H

 

 

 

 

H000 0000H 0000 0.0 0Hm0 000H HOH0 0H
0000 0000H 0000 00.0 0H00 000H m000 HHH
0000 0HO0H 0000 00.0 HOH0 HO0H 0000 HH
0000 0000 H000 00.0 0000 HO0H 0H00 H 000H
0H00 0000 0000 m0.m 0000 NO0H mmm0 >H
0000 0000 mHHm 00.0 0000 0HOH 0000 HHH
0000 0000 0000 m0.m H000 0000 0000 HH
0000 0000 0000 00.0 0HHO 0000 0000 H 000H
0000 0000 0000 00.0 0000 0000 0000 >H
N000 0000 0000 00.0 0000 000m 0H00 HHH
0000 H000 0000 00.0 H000 000H 0000 HH
0000 0000 0H0m m0.0 H000 mmmH 0000 H 000H
00H0 0000 0000 0m.m 0000 0H00 H000 0H
0H0 000 0000 00.6 0000 000H 0000 HHH
0000 0600 0000 00.0 00H0 0HOH 0000 HH
0 0000 H000 0H0m 00.0 00H0 HO0H 00H0 H 000H
14 0000 0000 0000 00.H 0000 NHNN HH00 0H
0000 H000 0000 00.H 0000 000H 0m00 HHH
0H00 H000 000m 00.H H000 000H 0m00 HH
0000 0000 0000 00.H H000 000H 00H0 H 000H
0000 0000 0000 00.H 0000 NO0H N000 0H
0000 0000 0000 00.H 0000 000H 0Hm0 HHH
00H0 00H0 000m 00.H 0000 0NOH HOH0 HH
0000 0000 0000 00.H 0m00 000H 0000 H 000H
0000 0000 H00m 00.H 0Hm0 0HOH 0000 0H
0000 0000 0H0m 00.H 0000 000H 0000 HHH
0000 0000 0000 00.H 00H0 000H 0000 HH
0000 00H0 N000 00.H 0000 000H 0000 H 000H
AmcoHHHHEV HmQOHHHHEv AchHHHHEV A000000QV AmcoHHHHEV AmuoHHHHEV AchHHHHEV 0000030
00002 002 000000 002 0000100002 0000 0900:00002 pmon 00H0 000002 0000 00 0:0
00 QOprom :00002 00 0 n 00>g000m
00:62 mnamm 0000
umschcools0 00000

 

.0 meme 00% mm 0500 "mmopsom
.mpHmoamQ mmcH>mm HmGOmpmm 0:0 mpHmoamm moHpoz pom 05mm
009 wH =m>HuomcH= mpwz pmzp mpHmome mo COHpLOQOLQ 00p MCHESmmm 0p 00meHpmm 000HI000H .0
.mpHmoamQ mwcHPmm Hmcompwm .o
.Humsv pqumye :H mampH pmHHom QMHUmcmo dea mpHmong 0cm mmpoz mnmcmo mo xcmm on
.mpHmoawD xcmm pmzpo .000H .HHH Eopw 00cm .mpHmome pCmECLm>00
HmHocH>ogm .mpHmoamQ 000200 mo pCmECLm>00 .mpc5000< wcHSUmno Hmcomhmm .mpHmoamU CCmEmQ .m

 

0QQOH 0HO0H 0000 00.0 0000H 0000 0000 0H
0000 0000H 0000 00.0 0000H 0000 0000 HHH
0000 0000H 00H0 00.0 00HOH H000 0000 HH
00H0 00o0H 00H0 00.0 00OOH H0H0 0000 H 000H
0000 0000H 0000 00.0 0000 0000 0000 0H
0000 0000H 0000 00.0 0000 00H0 00H0 HHH

1 0000 0000H 0000 00.0 0H00 0HHO 0000 HH

L. 0000 0000H 0000 00.0 0H00 000H H000 H 000H
0000 0000H HH00 00.0 0000 00H0 0000 >H
0000 0000H 0000 00.0 0000 0000 0000 HHH
0000 0H00H 00H0 00.0 HH00 0000 0000 HH
0000 0000H 00H0 00.0 0000 0000 0000 H 000H
00H0 o000H 0000 00.0 0000 0000 0000 0H
0000 000HH HHHO 00.0 0000 0000 H000 HHH
0000 000HH 0000 00.0 0000 000H 0000 HH
0H00 0HOHH 0000 00.0 0000 000H 0000 H 000H
0000 000HH H000 00.0 0000 00H0 0000 0H
0000 000HH H000 00.0 0HHO 000H 0000 HHH
0000 000HH H000 00.0 0000 000H 0000 HH
00H0 0H0HH 0H00 00.0 0000 000H 0000 H 000H
0000 omHHH 0H00 00.0 0H00 0000 0000 0H
0000 000OH 0000 00.0 0000 HO0H 0000 HHH
0000 0000H 0000 00.0 0000 000H 0000 HH
0000 0000H 0000 00.0 0000 000H 0000 H HO0H

 

'1

l‘.

 

 

#2

38

Conversations and logic, because of the low rate paid, seem
to indicate that they are small in number and dollar amount.

The second item in this classification is the non—
negotiable fixed term deposit receipts of $25,000 and over.
They are sold by the chartered banks to compete with the
Treasury Bills. They pay approximately the same rate as is
current in the money market and have maturities of from 30
to 36h days.39

The final type of debt instrument are negotiable term
notes issued in multiples of $10,000. The rates are period-
ically revised and the maturities run from two to six years.ho
3o Quebec Savings Banks

The Quebec Savings Banks are the remnants of the
"Savings Bank and Cooperative" movement of the 18003. Pre-
sently, there are only two operating in Canada, one in
Montreal and the other in Quebec City. They are authorized
by Dominion legislation and are regulated by Federal author-
ities.‘Pl

The Quebec Savings Banks produce pure money and money—
debto The money-debt is essentially the same as that produced
by the chartered banks. It is an interest bearing and check-

able deposit which is held primarily by individuals° However,

 

38With several members of the Research Department,
Bank of Canada, December, 19 7.

39Canadian Banker's Association, Submission, p. #9.

“OIbid. thoyal Commission, Report, p. lh7.

 

 

 

.-ve: in» :-

r., QWJ—F

©N.H
om.H
om.H
0N.H
©m.H
©N.H
©N.H
0N.H
mm.H
NNOH
5N.H
hm.H
mm.H
NN.H
5N.H
hm.H
0m.H
Om.H
Om.H
0m.H
Hm.H
Hm.H
Hm.H
Hm.H

HHNHH
HHHHH

Nr402
rar+H

OOO‘OOHOO‘KD‘OO O\O{\‘OOO\L\OO O\‘00
H

r-{

>H

HHH

HH

H dqu
>H

HHH

HH

H mqu
>H

HHH

HH

H NeoH
>H

HHH

HH

H HJOH
>H

HHH

HH

H oemH
>H

HHH

HH
H mmmH

 

AmcoHHHHEV
AmcoHHHHEV
kmcoz pmz

AmCOHHHHEV
pseudo umz awmcoz mo

cowppom mmcoz nflmm mpmm

Apcmommmv

Iwmcoz so

AmCOHHHHEV
mQOHHHHEV
I>mcoz p mcoz mesa

AmQOHHHHEV
m mm>nmmmm

 

 

mmz<m mOZH><m ommmsd

pmpmmso

Ho Cam

 

d mqm<e

an

mmH
mOH
00H
49H
mm:
omH
mmH
HmH
msH
msH
msH
05H
meH
eeH
m:
seH
moH
meH
00H
00H
mmH
mmH
mmH
QmH
04H
54H
msH
msH
emH
omH
emH
mNH
OHH
mHH
:HH
oHH

ONH
ONH
mNH
BNH
NHH

4N.H
JN.H
4N.H
:NoH
JN.H
dN.H
dN.H
4N.H
©N.H
©N.H
©N.H
©N.H
mN.H
mN.H
mN.H
mN.H
mN.H
mN.H
mN.H
mN.H
mN.H
mN.H
mN.H
mN.H
JN.H
4N.H
:N.H
JN.H
HN.H
HN.H
HN.H
HN.H
MN.H
MN.H
mN.H
mN.H

mam
sHm
mHN
sHm
mom
sow
How
com
mmH
smH
40H
soH
omH
moH
mmH
moH
mmH
mmH
smfi
mmH
Hen
Nefi
osH
osH
00H
00H
00H
mmH
osH
54H
meH
oeH
QmH
QMH
emH
mmH

HNHOOHHHHHNHHHHHHHHHHHHHHHHNHHOOOHHO

ON
ON
ON
ON
ON
mH
ON
ON
HN
NH
NH
0H
wH
5H
mH
NH
mH
MH
0H
ON
dH

mH
HN
mH
mH
0H
HH
MH
NH

NH
HH
mH
NH
sH

 

 

 

 

L5

 

mNH mom ssH sm.m mmm OH em >H
mNH OBN QJH dmoN me N ON HHH
MNH MON NdH dm.N NmN H ON HH
MNH mmN mMH :moN th m NN H 000H
mm oem HHH mmom men s om >H
00H mow HNH Nm.N Hum s mvH HHH
NOH mom HNH mm.m Hmm H cm HH
moH mom ONH mm.m onm m. ow H mmoH
mo com «HH mmom sew HH mm >H
mm mom oHH mmum mmm H mm HHH
mm 0mm NHH mm.m new 0 mm HH
so mmm ooH mm.m new m om H mHOH
em msm ooH sm.m «mm MH om >H
mm men moH swam com H mH HHH
em Hem moH sm.m 0mm H OH HH
mm mmm OOH sm.m mmm H mH H mmoH
NQH Nmm HNH esoH Hmm H Om >H
00H mmm mmH ee.H smm H 5H HHH
ooH 5mm NNH ee.H mam H NH HH
m0H mmm smH ee.H emm 0 OH H emoH
mo mmm eHH mo.H new 0 0H >H
HoH smm OHH mo.H mmm o mH HHH
em 0mm OHH we.H mew o OH HH
mo mHm HHH mo.H mmm 0 ON H mmoH
mo oom mHH mo.H emm H OH >H
mo mHm oHH mo.H mmm o mH HHH
om sow mHH mo.H emm H em HH
em 50H 00H mo.H on H mm H emoH
HmCOHHHHZV Apcwopmmv AmcofiSsz
HmsoHHHsz HmCOHHHHzl pawn ppmm pnmo AmcoHHHHeV HmcoHHHHsV empemsa
Hmcoz pmz pzapso Hmz a>msoz Ho Izmcoz so osmmcoz chocoz mesa mmm>mmmmm me new

:oHpHom Hoses UHmm mpmm

’ Illl'b

 

.emschcoo--s mHmHH

0HHH OHOOH .OmwOH .OpHmno>HcO
HHHOoz.mHmOOp .Hoz OOOmHHpsacsO HOOH-OOOH momcmo :H ammo: Ho OHOOOO OOH .opHmsz mHzcmO
.oocwbcoammppoo Hmcomnwm .mxcmm mo Hmpchw HouomQOCH .mnmcmo
.mocmvcoamwnpoo Hmcomnmm .mvmcmo mo xcmm .mvmsmo
aflmhmmz HmpmH new OOOH .mvmcmo Ho xcmmum30ppov nmEESm HOOHpmempm .wvmdmo Ho xcmm .mOmcmo
"d Osm m mmHnt pom mmopsom
pm meme .O OpoCPQOH mmm .U
=°mpsdooo< wcHSOmso Hmnomnwm: mmOdHocH .o
ompHmomwm pewscpm>ou HOHOCH>onm mzHQ mumcmo Ho pcmscpw>ow -.p
omxcmn Ompmupmco LpHS mpHmoamO msHQ mpHmomoo Ocm mmpoz mvmcmo mo xcmm nm

 

 

 

OON OH; OHN OO.N HNO OH ON >H
OOH NOH OHN OO.N ONO H NN HHH
HOH OOO OHN OO.N OHO H ON HH
OOH OOO OHN OO.N OHO O ON H OOOH
Au NOH Omm OOH OO.N NOO O ON >H
A. OOH ONO OOH OO.N OOO O Om HHH
OOH OOO NOH OO.N OOO N Om HH
OOH OOO OOH OO.N. HOO H mm H OOOH
OOH mmm OOH OO.N ONO OH ON >H
HOH Omm OOH mm.N ONO O OO HHH
HOH mmm OOH mm.N OOO N Om HH
OOH ONO NOH OO.N OOO H Om H OOOH
NHH ONO OOH OO.N rem O OO >H
OOH ONO OOH mm.N OOO O ON HHH
OOH OHM NOH OO.N OOH H ON HH
OOH OHO HOH OO.N OOO O ON H OOOH
OOH OOH OOH NH.N ONO OH ON >H
OOH OON OOH NH.N ONO N ON HHH
ONH OON OOH N4.N ONO H mm HH
ONH NON OOH NO.N OHO O Om H NOOH
OOH OON OOH OO.N OOO H ON >H
ONH OON HOH OO.N OHO N ON HHH
ONH OON OOH OO.N OOO O ON HH
ONH OON HJH OO.N OON O ON H HOOH

 

 

#7

it tends to have a little higher turnover rate than Personal
Savings Deposits.“2

The pure money produced consists of "Personal
Chequing Accounts" and Dominion and provincial government
deposits}*3 The data does not distinguish between money—debt
and the "Personal Chequing Accounts" for all years. For the
years that the division is available, the amount of "Personal
Chequing Accounts" is insignificant. Thus, no large biase
will be introduced by including all "Personal Checking Accounts"
in with money-debt. Thus, the pure money column will consist
of only Dominion and Provincial Government deposits.

The rate paid on money—debt and the charges on pure
money and money-debt are the same as for the chartered banks.MP
As with the chartered banks, it will be assumed that no ex—
plicit charges are made on any deposit and that the govern-
ment deposits pay no explicit rate of return.

A. Trust and Loan Companies

The Trust and Loan Companies had their beginnings in
the 19th Century cooperative movement which spawned the Quebec
Savings Banks. However, these two institutions have de-
veloped along very different lines. While the Savings Banks

have restricted themselves to dealings with individuals

 

h21bid., p. 150

43Bank of Canada, Research Department, Personal cor-
respondence.

thoyal Commission, Report, p. 150.

 

’mLfrl-fi:g:,g‘fl'L-_ ..- -. .- L— 7..

A8

in a certain geographic area, the Trust and Loan Companies
are concerned with general intermediation on a nationwide
scaleo

In addition to Estate and Trust Administration, the
Trust and Loan Companies have specialized in the intermediary
business which is prohibited for the chartered banks. They
are not bound by the maximum interest rate provision of the
chartered banks. Thus, one would expect that they are a
source of borrowings for those projects rejected by the
chartered banks. Also, the Trust and Loan Companies are the
major source of mortgage loans.

The Trust and Loan Companies produce pure debt, in the
form of Debentures, Guaranteed Investment Certificates, and
non-checkable deposits, and money-debt, interest bearing and
checkable deposits. The money-debt production is only a
small prOportion of total "liabilities", 17% in 1966.l*5
Thus, their net money production is probably underestimated
to a greater extent than that of producers with a smaller
prOportion of non-money output.

The money-debt of the Trust and Loan Companies is
similar to that of the chartered banks. It is mostly held by

individuals and the companies maintain an active policy of

 

h50heckable deposits as a percentage of checkable
plus non—checkable deposits plus Debentures plus G. I. 0.3.

 

49

discouraging high turnover accounts.46 However, the average
size of an account is three times that of Personal Savings
Deposits and the rate of turnover is slightly larger.“7 The
little information available concerning the actual procedure
of interest and cost calculation seem to indicate that they
are similar to those of Personal Savings Deposits.A8

Complete information concerning Trust and Loan pro-
duction is not available for all years. Only since 1963
has the Dominion Bureau of Statistics distinguished checkable
from non-checkable deposits. From 1951 to 1963 only total
deposits, on a quarterly basis, were reported. Prior to
1951, only yearly reports are available.

In order to estimate the quarterly breakdown of
deposits for the 1951-1963 period, a sample of companies repre—
senting 68% of the total assets of Trust and Loan companies
operating in Ontario was used.l+9 It is assumed that the
ratio of checkable to total deposits for this sample is the
same as for all Trust and Loan Companies operating in Canada.
This sample seems to be biased toward large companies.50
Thus, if large companies tend to have more checkable deposits,

the estimates will overstate the true money - debt

 

46Royal Commission, Report, pp. 178-181.

47J. Fo Graham, gp_§l.,.fiplg, p. 11—53.

h8Royal Commission, Rgpppp, p. 182.

1+9J. F. Graham, e_t_al_._, 393,, p. 11—58. 50%” pm

 

 

50

production.51

The Trust Companies have always accepted non-checkable
deposits. The Loan Companies began to offer and the Trust
Companies emphasize non-checkable deposits in response to the
growth of the short term money market in the mid 1950's.52
These deposits are designed particularly to compete with the
"Other Notice Deposits" of the chartered banks. From 1951
to 1955, before their growth began, non-checkable deposits
were consistently equal to 15% of total deposits. Conse-
quently, checkable deposits, money-debt, for the 1939-1950
yearly series will be estimated by calculating .85 of total
deposits.

The only available interest rate information is the
actual rate paid on total deposits for companies operating
in Ontario and, for the 1951-1966 period, the stated rate of
checkable deposits. The rates paid on total deposits will
diverge from the rate paid on money-debt as the ratio of
moneyedebt to total deposits declines. For the 1939-1950
period, when the ratio of checkable to total deposits is
assumed to remain constant, the actual rate paid on total
deposits to the stated rate on money-debt, was .83. Thus,

the actual rate paid on money-debt will be assumed to be

 

51Judging from the company data available from the
Registrar of Loan and Trust Corporation in Ontario, large
companies are more likely to have deposits, the type of
deposit is not known, than are small companies.

52J. F. Graham, et al., Role, pp. 11-23.

 

 

 

 

 

 

OO NmH , Om mm NO.H HNN >H
OO OON ON OO NO.H ONN HHH
HO OOH Om HO NO.H mNN HH
Om OOH Om mO NO.H mmN H HOOH
mmupmsa
Ho Ocm
Om OOH mm OO HO.H OON OOOH
OO OOH Om OO OO.H OOH OOOH
Om NOH Om OO OO.H NOH OOOH
.1 NO ONH Om NO OO.H OOH OOOH
.3 mm OOH ON NO OO.H HmH OOOH
HN Om ON OO OO.H OOH OOOH
mH mO NN 0O OO.H OO OOOH
OH OO OH Om OO.H OO mOOH
OH Om ON Om OO.H OO NOOH
OH OO OH mm OO.H OO HOOH
NN HO OH OO OO.H OO OOOH
O NN HO OH OO OO.H OO OOOH
O
, AOCOHHHHEV HmcoHHHHEv AmcoHHHHEV AmcoHHHHEV Anamohmav AOCOHHHHEV pmmw
Omcoz pmz pumpso pmz o.mmm>pmmmm meQIOmdoz ppmo O.mpnmonOmcoE
mo :oHupom n>msoz so
Omcoz QOHmm mpmm

 

mmHz¢m200 z¢OH Dz< Ombme
m mHm<B

 

52

OOm
HNm
MHm
OON
OON
HON
HwN
OON
OON

HON
OON
OON
OON
HON
NON
OHN
OON
NON
OHN
HHN
OON
OOH
OON
OOH
mOH

HMH
ONH
OHH
OHH
OO
OO
OO
OO
OOH
NOH
mOH
OOH
OHH
OHH
OOH
OOH
mHH
NHH
OOH
OO
OOH
OOH
OOH
mOH
wO
OO
OO
OO

o s O

N\O\O\O\O\O\O\‘O\O

o

NNNNNNNNNNN

100013330000 MMNTD
NNN

c
N

HH.N
HH.N
HH.N
HH.N
OO.H
OO.H
OO.H
OO.H
mO.H
mO.H
mO.H
mO.H
OO.H
OO.H
OO.H
OO.H

mmm
mwm
Omm
Omm
ONm
Hmm
ONm
OHm
OHm
OOm
OHM
HNm
ONM
Omm
mNm
OOm
OON
HON
MON
OON
mON
mON
NON
OMN
Nmm
mNN
ONN
ONN

>H
HHH

H mOOH

 

 

 

WEW.;§.'~.~,

 

53

 

OmH OOO HO mNN mO.m OOm >H
OOH OOm HO ONN mO.m mOm HHH
mHH OOO mO OON mO.m OOm HH
OOH NNO OO OOH mO.m NHm H OOOH
OO NHO NO HOH OO.N OOO >H
NNH NmO Om OOH OO.N OOO HHH
mOH OHO OO HOH OO.N mOO HH
OOH OHO OO OOH OO.N OOO H NOOH
mHH OOm NO mOH OO.N OOO OH
ONH mOm OO OOH OO.N mmO HHH
OHH OOm Hm OOH OO.N mNO HH
mNH OOm mO OOH OO.N ONO H HOOH
OOH Omm Om mOH mO.N OOm >H
HHH Omm Hm NOH mO.N OOm HHH
OOH mNm mO OmH mO.N OOm HH
OHH ONm OO OmH mO.N OOm H OOOH
Om OON mO mNH OO.N Omm >H
mO OOm Om OmH OO.N HOm HHH
OO Omm mm ONH OO.N NOm HH
mO OOm Om HmH OO.N mOm H OmOH
HOCOHHHHEV HmcoHHHHEO HOCOHHHHEO AmcoHHHHEO Apcmopmmv HOOOHHHHEO pmupmsa
Omcoz pmz psmpso pmz mm>gmmmm pmenOmcoE pnmo O.mpan|Omcoz mo Omm
mo :oprom IOocoz so
Omcoz Ome mumm

p

 

.OOOSHQCOOIIm MHmde

 

 

.HmhmmO ”mpmH Omm OmOH
.pmchpm m.cmmso "oucopoev ppommm .mchmQEoo pmdpe Omm OOOH mo ampumHOmm gngmuzo

, OOO-HH 0am .HOOOH .oHpmpco

cumumm3 mo OuHmpm>HOO .COHpmppchHEO< mmmchsm mo HmowvmuCOOQOHV NEocoom
cmHOmcmO OOO :H OOHOOQEOO OOOH Ocm pOSLO asp mo mHom 029 ..HO pm .Emsmpo .O 0O
.mocmOcogmmppoo Hmcompmm .pcmEpprmO Oopmmmmm .mOmcmO mo xcmm .mOmcmO

. mammO
memH OOO mOOH .mOmcmO Mo xcmm "OBOpQOV Opmasdm HOOHpmempm .mOmcmO We xcmm .mvmcmo
«mmopsom
.pxme Omm .NOOH OH on HmOH H pom prmEHpmm .O
.Oocmphdo

cmepom mEow Omm .mpHmoamO xqmn Ompmppmzo .mpHmong Omm mmpoz mnmcmo mo xcmm oo
.COHpm>Hme pom pxmp mmm .O
.OHpmpCO OH Ochmpmmo mchmacoo OHco mmudHocH mOOHIOmOH .m

54

 

 

mOH HHO ONH MON Nmom HMO >H

OOH NNO OOH MON Nmom HmO HHH
OmH OHO ONH OON Nm.m OMO HH

ONH OOm mOH NON Nm.m mOO H OOOH
OOH OOm OmH OmN NN.m HNO >H

HmH OOO mOH OmN NN.m mHO HHH
OOH OHO mHH NON NN.m NmO HH

mOH OOm OO 4 mON NN.m OOO H mOOH
OO OHm OOH mON OO.m OOO OH

OOH OHm ONH MMN OO.m OOO HHH
OmH mNm OO ONN OO.m mHO HH

mOH OHm OO OHN OO.H OOm H OOOH

 

 

 

55

.83 of the stated rate for the 1951-1966 period.53

5. Credit Unions
The Canadian credit unions may be divided into two
groups according to how they view their function. The French
speaking Caisses Populaires think of themselves as a "savings
cooperative" which will make loans to its members only under
special circumstances.54 On the other hand, the English
speaking credit unions View their function as one of pro-
viding low cost loans to its members for any of a variety
of reasons.55
The difference in philosophy has led to a marked
difference in the product mix offered by the two types of
credit unions. The English speaking credit unions rely
mostly on shares and non-checkable deposits, with checkable
deposits available only in certain areas or from a few large
credit unions, as a source of funds. The sales of the
Caisses Populaires are almost entirely made up of money-
debt with shares often being limited to one per member.56

The difference in the types of "deposits" offered by

 

53The implicit assumption is that the ratio of mini-
mum to average balance remains constant over the period. The
data for the chartered banks covering the same period seem to
indicate that this assumption is justified.

5“La Federation Des Caisses Populaires Desjardins,
Brief Submitted by La Federation de Quebec des Unions
Re ionales des Casses Po ulaires Desjardins to t e oyal ng-
mission on Bankin and Finance (Levis: La Federation Des
Caisses F i

opulaires Desjardins , p. 39.

55Royal Commission, Re ort, p.155. 56Ibid., p. 159.

   

 

56

the credit unions has led to difficulties in estimating their
money production. The only nationwide data avaiable for
credit unions does not distinguish between checkable and non-

checkable deposits.57

Consequently, each region was examined
individually to isolate checkable deposits.

Most credit unions and all Caisses Populaires belong
to one or more of 18 voluntary regional leagues.58 Correspond-
ence and examination of publications has disclosed that two
of these associations have records which provide information
on the checkable deposits of its members. Luckily, these
two leagues accounted for 92% of all "deposits at credit
unions" in 1966.59 Increased coverage would require an
examination of each individual credit union to see if its
"deposits" are checkable.

The decision as to the interest paid on checkable

60

deposits is made by the individual credit union. Consequently,
some credit unions produce pure money while others produce
money—debt. Interest payments are available only for an entire
leagUe. Thus, it will be assumed that each unit sells money-

debt and an average rate on the checkable deposits for the

 

57Canada, Department of Agriculture, Credit Unions in
Canada (Ottawa: Department of Agriculture, l9h5 and later
years .

58Royal Commission, Report, p. 165.

59The 1966 MoneyéDebt entry of Table 6 as a percent of
"deposits at credit unions" in the source cited in 57.

60Brief of La Federation, p. 110.

 

57

entire league will be computed. Among the Caisses Populaires,
interest is computed on minimum quarterly balances.61 No
information is available concerning the charges made on
chequable deposits.

Each of the leagues used in this study has a credit
union "central." The "central" is a combination investment
agency and "bank of last resort." The local societies are
encouraged to deposit their "excess" funds in the "central"
and to borrow from it when necessary.62 The deposits of the
locals in the "centrals" are not only an interest bearing
asset but are also used for check clearing and other inter-
local transactions. It may be argued that, given the credit
unions' strong preference for "liquidity," if they did not
have the "centrals" many of these deposits would be in the
chartered banks. The question becomes one of deciding if all
the "reserves" of the local, deposits at "centrals" plus
checkable deposits in other institutions plus cash, or the
consolidated position of the league, cash held by the
central plus cash and checkable deposits of the locals,
should be subtracted from total money output in order to
obtain net money production. Unfortunately, a choice is
available only for recent years because the consolidated

cash positions are not available for 1950. The decision

 

E)llbido

62Royal Commission, Report, p. 165.

 

 

 

0/

Ex/

 

OOHupom Omcoz

0.9eHmm mpmm

Hommmsov monZO OHQmmO

O mHmde

H.mOO H.OON 0.00N MH.N OomNm 0.00 OmOH
m.MNO m.OOH O.mON OO.N 0.0mO H.Om OmOH
0.00m N.NOH H.mOH OO.H moHOO 0.0m OmOH
m.ONm O.HHH OoOMH NOOH OoOOm 0.0N mmOH
0.00N 0.00H 0.0NH HO.H OoOOm 0.0H .OmOH
0.00N OomO OoHHH OO.H OOOON 0.0H mmOH
moONN OoOO OoOO mO.H OoOON 0.0N NmOH
m.OON m.Om moOO OO.H OOOHN NoOH HmOH
OoOOH OoOH OoOm NO.H moOOH OOOm OmOH
O.mmH 0.0N Oon NOoH OOOOH m.Hm OOOH
OOHOH Oomm OOOm OO.H HomOH mon OOOH
0.0NH O.Nm 0.0m OO.H OommH OoOm OOOH
OoNOH OoON O.Hm OO.H OoONH OoON OOOH
OomO OOOH OoOm OO.H O.HO 0.0H mOOH
O.Mm m.OH 0.0N OO.H HOOO m.mH OOOH
H.Hm 0.0 0.0H HO.H OOHO 0.0H mOOH
OoOH O.m 0.0H mO.H OomN 0.0 NOOH
NomH H.N N.O mO.N moOH H.m HOOH
O.HH m.N O.m OH.N OoOH mom OOOH
0.0H O.N N.m mN.N OomH OON OmOH
HmcoHHHHEV AOOOHHHHEV AOOOHHHHEV Hpcmopmmv HmOOHHHHEV HmcoHHHHsv pmmw
Omcoz pmz psapdo pmz uQmO ppmaaOmcoS 0.9.mmm>pmmmm
IOmcoz mo aOmcoz co

 

 

eOO.HOOO nOpOH OcO OOOH .OOHpmeOpO mo sOopsm
omposa "OpHO OOQOSOV HOOOOO HOOHpmeOpO OODOSO .OOHpmeOpm mo SOOpsm ompmso .OOQODO
‘ .> OHOOO .HH XHOCOQOO ANOOH .OocOcHO OOO
OCHOOOO co coHOOHEEoO HOOom "OzOpva OocOcHO OOO OOHOCOO co :onmHEEoO HOOom
OOH pom OOLOQOLO LOQOQ OOHOLQB OOOLHOHSQOO mOmmHOO OOO OOOHEO pHOOhO .Opdohmz OOHHO
.OocOOconOppoO
HOcompOm .pOOEppOQOO LopOOmOm .OCHOOOOOOO OOLHOHOOom mOmmHOO OOO OOHpOpOOOm OH
.mmsOm OOO .AOOOH .OQHOLOOOOO OOLHOHSQom OOOOHOO
OOO coHpOLOOOO OH "OH>OHV OpHoEOz .OQHOOOOOOO OOLHOHSQom mOmmHOO OOO OOHpOLOOOm OH
.OOH- mH tag ..NHOH .mcHepOHmmo mmpHmH
usaom mOmmHOO OOO :oHpOeOOOO OH "OH>OHO.O0OO:HO OOO CHOCOm so conmHEEoO
HOOoO OOO op mcHOpOOOOO mOpHOHOQom OOOOHOO OOO OOHOcoHOOm msoHOD OOO OOQOSO OO
coHpOnOOOO OH Om OOppHEOSO OOHLO .OOHOLOOOOO OOLHOHSQom mOmmHOO OOO coHpOpOOOm OH
homHOOO LOOOH OCO OOOH .OLSOHSOHLOO
Oo OCOEOLOOOO "OzOppOV OOOOOO CH mcoHCO OHOOLO .OOOHHSOHLOO Oo OOOEpLOQOQ .OOOOOO
"mOoLdom
.Om OCOO co OOO op OOESOOO mH OOOOHOO HHO mo LOOO HOomHO OOO .o
.OOHOLOOOOO OOLHOHOQOO OOOOHOO OHco mOOSHocH .OOOHINOOH .9
=.mHOppcOo: cH mOmmHOO HOooH Ho OpHmoOOO OOOOHocH .OmOHnOmOH .O

 
 

 

O.HHOH m. mOm H.HmO OO.N H ONHH O.OOH OOOH
m. mOO 0.0ms O.mmm OO.N 0 OOO O.mO mOOH
O. OOO m.Omm m.OHO Om.m O OOO 0.00 OOOH
O.ONO o.mOm H.NOO mO.N m.OOO m.OO HOOH
m.mOO 0.0Hm N.OOm HN.N O.mmO m.OO NOOH
0.0HO O.mHm 0.00m NH.N 0.000 0.00 HOOH
O.mOm O.HON 0.0mm OH.N OQONO 0.0m OOOH
m.NOm 0.0mm m.OON OH.N OOOOm NamO OmOH

 

61

III. The Money Supply Series

The information provided in Tables 2 through 6 may be
used to construct a number of different money series. The
"net output" and the "net money" entries of each producer may
be combined into parallel series differing only by the debt
component of money-debt. The summation of the net output of
each producer would then correspond to what may be called
a "traditional" concept of money. In other words, money—
debt is included in the supply of money at its nominal
value. The use of the net money entries allows the
construction of a money series based on the analysis presented
earlier, i.e. the debt component has been estimated and elim-
inated° These two series may then be used to test the
validity of implications derived from the hypothesis pre—
sented in Section I.

Unfortunately, equivalent data is not available from
every producer for each time period. Consequently, a number
of pairs of series were constructed differing in c0verage
and time span. Each pair is presented and discussed below.

The basic quarterly series, Q and Qa’ runs from

u
the first quarter 1939 to the fourth quarter 1966, and in-
cludes the production of the government, chartered banks,
and the Quebec Savings Banks. Unfortunately, the comparison
rates and the rates paid on money—debt, used to estimate

the debt component of money—debt, could be computed only on

a yearly basis. Consequently, Qa, the series from which the

 

62

debt portion has been eliminated, is not a truly independ-
ent quarterly series. Its time trend within each year should
be identical with that of Qu, the unadjusted series. The
major discontinuity in Qa would appear between the fourth and
the first quarter, when the new money-debt adjustment factor
is applied) In order to minimize this discontinuity and to
facilitate graphical comparisons, both Qu and Qa were
seasonally adjusted by means of repeated applications of
a ratio to moving average method. The seasonally adjusted
series are entitled Q3 and Q3.

Figure 1 presents Q3 and Q: in graphical form along

63

with reference cycle turning points. As expected, the
levels and the time profiles of the two series are almost
identical“ The fourth to first quarter discontinuities in
Q: cause the sign of the rate of growth of this series

to differ from that of Q: in seven of the twenty six op—
portunitieso However, only two of this, from the fourth
quarters of l95h and 1957, are of significant magnitude. In
both instances the rate of growth of Q: is positive while
that of Q: is negative and this change in sign preceeds a
trough in the reference cycle by about one year. However,
Figure 2 dispels any hope that Q: may be a better cyclical

, S . . .
predictor than Qu. The two instances mentioned above prove

 

63Keith Hay, "Money and Cycles in Post Confederation
Canada," Journal of Political Economy, 75 (June, 1967 , a
263w273.

 

   

 

 

20:09:30 :4 3:4 non—Em
39:0 82 Pa >952 62

H mass“.

8302.80 2.8.63 2c. «Ste. 2325

“.2250
33:0 32 m .382 32 go 3525 go 2st

 

 

65

to be simply extentions of an already declining rate of growth
with the troughs in Q: coinciding with those of Q3.

The similarity of the two series is more strikingly
demonstrated in Figure 2 where centered five quarter moving
averages of the rates of growth of Q3 and Q: are plotted.

The cyclical profile, with the exception of the downturns

in l9h8, 195h, and 1958, and the turning points of the two
series are almost identical. Thus, the discovery of differ-
ences in the cychcal behavior of the two series, if they
exist, must await more sophisticated analytical tools.

Another quarterly series, running from the first
quarter 1951 to the fourth quarter 1966, was computed in
order to include the quarterly data of the Trust and Loan
Companies. The estimated net output and money production of
these institutions, presented in Table 5, is small and
probably would cause little change to the previous series
if it were simply added when it became available.6h In the--
possibly vain—-hope of providing data as free as possible from
exogenous influences, separate series, Qfi and Q; and their
seasonally adjusted counterparts Qfis and Qés, were constructed
which cover the period when quarterly Trust and Loan in—
formation is available. No separate analysis of these series

was attempted because the small increase in coverage should

 

6brThe net output and net money of the Trust and Loan
Companies is probably underestimated to a greater degree than
that of any other producer. These companies are not as insig-
nificant and the data in Table 5 seems to indicate. See
above, pp. h7~5ho

 

 

 

66

have little effect on cyclical behavior.

The use of annual average net output and net money
production will enable series to be constructed which will
incorporate data from all producers in every time period.
The yearly series, Mu and Ma, run from 1939 to 1966 and,
because of the inclusion of the credit unions, constitutes
a major improvement in coverage. Also, the corresponding
entries of the two series are independent in the sense
that the money-debt adjusting factor differs in each
period.65 Thus, the differences between the two series
should be capable of detection with relatively simply

statistical tools.

 

65This is reflected in the fact that the simple
correlation between the rates of change of Mu and Ma is
.335 while that of Qu and Q8 is °856.

 

 

 

 

QUARTERLY NET OUTPUT (Qu)
(millions)

(O

V

TABLE 7

g

 

 

 

Quarter I II III IV
Year

1939 2533 2533 2656 2839
1940 2822 2810 2876 2868
1941 3108 3238 3303 3313
1942 3617 3363 3783 39l8
1943 4117 4476 4480 4765
1944 4861 5236 5304 5721
1945 5642 6225 6008 6780
1946 6730 6622 6830 7036
1947 7105 7007 6998 7231
1948 7230 7486 7660 7929
1949 7912 8148 8456 8189
1950 8375 8384 8637 8786
1951 8713 8694 8738 8777
1952 8800 8853 9098 9227
1953 9426 9551 9616 9783
1954 9752 9816 10083 10213
1955 10342 10732 11029 11285
1956 11191 11405 11430 11297
1957 11364 11380 11524 11824
1958 11660 12152 12751 12949
1959 12937 13052 12909 12963
1960 12834 13081 13141 13676
1961 13451 13449 13885 14601
1962 14509 14808 14524 15075
1963 14735 15301 15412 16285
1964 16135‘ 16422 16370 17005
1965 17040 17567 17874 18625
1966 18279 18629 18975 19620
Sources:

Tables 1 to 6.

 

 

68

. TABLE 8
QUARTERLY NET MONEY (0a)

 

(millions)

 

 

 

Quarter I II III IV
Year

1939 1957 1985 2085 2249
1940 2257 2266 2316 2313
1941 2511 2732 2764 2730
1942 3072 2805 3172 3342
1943 3422 3824 3743 4043
1944 4010 4396 4353 4786
1945 4579 5191 4852 5649
1946 5428 5253 5399 5619
1947 5556 5437 5391 5613
1948 5415 5673 5799 6057
1949 5862 6067 6341 6070
1950 6117 6140 6394 6550
1951 6636 6628 6649 6682
1952 6797 6841 7051 7158
1953 7402 7501 7518 7763
1954 7155 7140 A7338 7482
1955 7360 7668 7864 8189
1956 8052 8221 8173 8049
1957 7474 7461 7510 7876
1958 7735 8094 8487 8747
1959 8722 8787 8623 8877
1960 9202 9408 9383 9966
1961 9467 9428 9771 10544
1962 10260 10507 10157 10806
1963 10259 10754 10736 11694
1964 11311 11570 11389 12073
1965 12012 12488 12621 13380
1966 13164 13470 13686 14404
Sources:

Tables 1 to 6.

 

“*i*'.m#i-q‘s Ell-'fi'aoV '_“""..w :aw ".~ ~.

 

 

 

TABLE 9

 

 

QUARTERLY NET OUTPUT, SEASONALLY ADJUSTED (Q3)

(millions)

 

 

 

 

 

 

 

 

 

 

 

Quarter I II III IV
Year

1939 2675 2808
1940 2826 2815 2897 2837
1941 3112 3243 3327 3277
1942 3622 3369 3811 3876
1943 4123 4484 4513 4714
1944 4868 5245 5343 5660
1945 5650 6236 6053 6707
1946 6740 6634 6881 6961
1947 7116 7019 7050 7154
1948 7249 7499 7717 7844
1949 7924 8162 8519 8101
1950 8388 8399 8702 8692
1951 8726 8709 8803 8683
1952 8814 8869 9166 9128
1953 9441 9568 9688 9679
1954 9767 9833 10158 10104
1955 10358 10751 11112 11165
1956 11208 11425 11516 11176
1957 11382 11400 11610 11698
1958 11678 12174 12846 12811
1959 12957 13075 13006 12825
1960 12854 13104 13239 13530
1961 13472 13473 13989 14445
1962 14532 14835 14633 14914
1963 14758 15328 15527 16111
1964 16160 16451 16493 16824
1965 17067 17599 18008 18427
1966 18308 18663

Sourges:

Tables 1 to 6.

 

 

70
TABLE 10

 

 

QUARTERLY NET MONEY, SEASONALLY ASJUSTED (03)
(millions)

 

 

 

 

 

 

 

 

 

 

Quarter I II III IV
Year

1939 2119 2200
1940 2270 2267 2354 2263
1941 2526 2733 2809 2670
1942 3090 2806 3224 3269
1943 3442 3825 3804 73955
1944 4034 4398 4424 4682
1945 4607 5193 4931 . 5526
1946' 5461 5255 5487 5497
1947 5589 5439 5479 5491
1948 5448 5675 5894 5926
1949 5897 6069 64445 5938
1950 6154 6142 6499 6408
1951 6676 6631 6758 6537
1952 6838 6844 7167 7003
1953 7447 7504 7641 7597
1954 7198 7143 7458 7320
1955 7405 7671 7993 8011
1956 8101 8224 8307 7875
1957 7519 7464 7633 7705
1958 7782 8097 8626 8557
1959 8775 8791 8765 8685
1960 9258 9412 9537 9750
1961 9524 9432 9931 10316
1962 10322 10511 10324 10572
1963 10321 10759 10912 11441
1964 11380 11575 11576 11812
1965 12085 12493 12828 13090
1966 13244 13476

Sources:

Tables 1 to 6.

 

”'9"? ’.V' L - 3'" ' . -

 

 

'L— “:27’57‘fr'. ‘ ," :v-f—‘V .r - ~ ., :._' _f;;‘-' KNEE -— .- 'V " ‘V~ “3‘ .J '- 7 q a H "1"”:

 

71
TABLE 11
QUARTERLY NET OUTPUT AND NET MONEY, INCLUDING TRUST
AND LOAN COMPANIES (Qfi and Q5)

 

 

 

 

 

 

 

 

 

 

Net Output

(millions)

Quarter I II III IV
Year

1951 8910 8892 8938 8959
1952 8993 9052 9301 9412
1953 9632 9762 9830 9985
1954 9958 10029 10325 10454
1955 10606 11006 11317 11556
1956 11476 11683 11700 11555
1957 11643 11661 11815 12092
1958 11950 12465 13072 13277
1959 13284 13391 13253 13257
1960 13162 13404 13478 14013
1961 13830 13823 14280 14995
1962 14925 15227 14956 15487
1963 15157 15750 15916 16783
1964 16649 16947 16880 17523
1965 17619 18186 18482 19193
1966 18839 19239 19597 20231
W
Net Money

(millions)

Quarter I II III IV
Year

1951 6693 6689 6715 6731
1952 6859 6910 7123 7209
1953 7473 7575 7594 7827
1954 7212 7195 7411 7549
1955 7421 7728 7934 8244
1956 8122 8292 8236 8094
1957 7528 7517 7568 7914
1958 7801 8167 8552 8819
1959 8815 8881 8716 8935
1900 9316 9517 9494 10074
1901 9590 9545 9891 10657
1962 10369 10612 10279 10905
1963 10365 10867 10885 11828
818 E89 898 13122
1966 13293 13629 13860 14567

 

(Sources:
Tables 1 to 6.

QUARTERLY NET OUTPUT AND NET MONEY, INCLUDING TRUST

72
TABLE 12

AND LOAN COMPANIES, SEASONALLY ADJUSTED

(Qfis and Qés)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Net Output

(millions)

Quarter I II III IV
Year

1951 6827 6565
1952 6913 6921 7242 7032
1953 7531 7587 7721 7634
1954 7268 7207 75351 7363
1955 7479 7741 8066 8041
1956 8185 8306 8373 7895
1957 7587 7529 7694 7719
1958 7862 8180 8695 8602
1959 8884 8896 8861 8715
1960 9389 9533 9652 9826
1961 9665 9561 10056 10395
1962 10450 10629 10451 10632
1963 10446 . 10885 11067 11537
1964 11543 11725 11688 11870
1965 12252 12658 12985 13155
1966 13397 13652

Net Money

(millions)

Quarter I II III IV

Year

1951 8985 8848

1952 9844 9067 9350 9296

1953 9686 9778 9881 9862

1954 10014 10046 10379 10325

1955 10666 11024 11376 11413

1956 11541 11703 11761 11412

1957 11709 11681 11877 11943

1958 12017 12486 13141 13113

1959 13359 13414 13323 13093

1960 13236 13427 13549 13840

1961 13908 13846 14355 14810

_1962 15009 152531 15035 15296

1963 15243 15777 16000 16575

1964 16743 16976 16969 17307

1965 17719 18217 18579 18956

1966 18946 19272

36urces:

Tables 1 to 6°

 

TABLE 13

YEARLY NET OUTPUT AND NET MONEY (Mu AND Ma)

 

 

 

 

 

 

 

 

Year Net Output (Mu) Net Money(
(millions) (millions?)

1939 4373 2717
1940 5263 2916
1941 6503 3313
1942 8337 3745
1943 9043 4551
1944 9826 5412
1945 9840 6320
1946 9821 5487
1947 10361 7336
1948 12003 7860
1949 12905 8488
1950 14161 8883
1951 16588 9125
1952 18654 9419
1953 19294 10062
1954 19032 10477
1955 20737 11441
1956 23166 11975
1957 24011 12226
1958 25011 13176
1959 26482 13838
1960 27433 14087
1961 28250 14875
1962 30653 15814
1963 32869 16628
1964 35397 17800
1965 38919 19273
1966 43306 20487
Sources:

Tables 1 to 60

 

 

 

CHAPTER III

SOME EMPIRICAL ISSUES
I. Introduction

The central hypothesis of this study is that only a
portion of the value of checkable deposits that earn explicit
interest should be considered as money. The implications of
this analysis were derived only insofar as it was necessary
to construct specific money series. The novelty of this
approach, and the time constraint placed on the author, has
not allowed inferences to be made which would enable a
direct test of this hypothesis. However, it was felt that
some of the statistical properties of the data gathered
should be presented.

The demand for money was chosen as a topic which
could display the attributes of the constructed money series.
Consequently, several classical money demand specifications
are fitted to money series derived in Chapter II. However,
in this chapter, precise theoretical analysis is kept
to a minimum. Results which, on an intuitive level, sup-
port the hypothesis are noted but will not be examined

in detail. Other findings which bear on more general

74

 

 

75

topics in monetary analysis will also be discussed. Thus,

this section is intended to be simply a presentation of
characteristics of the constructed series which might be
worthy of future study and not a rigorous test of the pre-

ceeding analysis.
II. The Demand for Money in Canada

A. General Comments

The demand for money may be viewed as a partial
equilibrium model or as one of the many interdependent
markets of the economic system. In the single equation
model, it is assumed that the actual money stock is equal
to the demanded stock and then the money series is regressed
against in number of variables which are thought to repre-
sent magnitudes which influence the demand for money. The
regression coefficients are then interpreted as represent-
ing the parameters of the demand for money function.1
The same empirical specification of independent and depend—
ent variables may, according to a different model, describe
a reduced form of a sector of the economic system. The
estimated parameters are then subject to an entirely differ-
ent interpretation. This possibility of error in speci-

fying the underlying model should be kept in mind when

 

1It may be argued that, on a macro level, the train
of causation runs from money to the "independent" variables.
See A. A. Walters, "The Demand for Money-The Dynamic Proper-
ties of the Multiplier," Journal of Political Economy, 75
(June, 1967), 293.

 

76

interpreting the results which will be presented.

The various series of Chapter II were constructed
so as to facilitate testing of the treatment of money-debt.
Corresponding series are identical except that the debt
component is eliminated in one but included in the other.
Thus, the various specifications may be run using both
series and the results compared to see if they shed any
light upon the validity of the hypothesis underlying
their construction.

Assuming that the analysis of Chapter II is cor-
rect, that interest bearing and checkable deposits are not
entirely money, it is not at all clear what differences should
be expected between the regressions of the independent vari—
ables on the Net Output and the Net Money series. The Net
Output series supposedly contains a non—money, a pure debt,
component. pBut debt is probably influenced by the same
factors, income, prices, and the interest rate, that affect
the demand for money. Thus, unless further information
can be obtained concerning the effect of the independent
variables on the demand for debt, very little can be said
about the differences between the Net Output and the Net
Money regre551ons. All money demand specifications were
run for the Net Output and the Net Money series and the
results will be presented side by side. However, the
differences in the results will not be discussed unless an

intuitive explanation can be given.

 

77

The usual specification for the demand for money has
income or wealth, in real or nominal form, as the scale vari-
able and the interest rate representing the alternative
costs of holding money. Various simple formulations involving
these variables will be run using several of the series de-
rived in Chapter II.

Net National Product, in nominal and deflated terms,
was chosen as the appropriate counterpart to "income." All
of the income series available from the Dominion Bureau of
Statistics are highly correlated and give almost identical
results. Net National Product was selected because its
definition seemed to correspond most closely to the theoreti—
cal concept of income. Current income is available in both
yearly and, beginning in 1947, quarterly form. In order to
facilitate comparisons, most specifications will be pre-

sented using quarterly, yearly, and permanent income data.

B. The Interest Rate

 

The alternative costs of holding money can best be
described with an index of all interest rates. In the absence
of such an index, individual rates are included in the
specification in the hope that their characteristics conform,

more or less closely, to that of the entire spectrum of rates.2

 

2David Laidler, "Rate of Interest and the Demand for
Money," Journal of Political Econom , LXIV (December, 1966),

547.

‘- yvr

 

 

 

in

I“.

Ix,

II!

it

3‘

n.

 

78

On the basis of a priori knowledge, one is unable to state
which of a number of different rates would perform best in
the various specifications. Consequently, a number of long
and short rates on both government and private bonds were
included in the various models in this study.

None of the interest rates chosen gave consistently
significant results in the quarterly formulation of the de-
mand for money. The only rate which seems at all related to
the money series is the end of quarter rate on three month
Treasury Bills. However, the coefficients yielded by the
current value of this rate, presented in equations Q1 and
Q2, are either insignificant or of the "wrong" sign.

It seems plausible that a quarter may be too short
a time span for money balances to be adjusted for changes in
the interest rate. Consequently, a first difference and
percentage change specification, used to eliminate multi—
collinearity, of the parameters in the current and the
previous time period was run, and the results are presented
in equations Q3 and Q4° As in the previous results, the cur-
rent value of the interest rate yields insignificant results.
However, the values of the interest rate parameters lagged
one period to show a significantly negative coefficient.
Thus, the results would not be inconsistent with a hypothesis
which stated that, due to either a lack of information or the
difficulty and costs of altering a portfolio, the demand for

money adjusts to interest changes only with a lag of several

 

79

time periods.

In contrast to the interest rate, changes in income
do seem to have an immediate impact on the demand for money.
In three of the four multi-period specifications, the sign
of the current income parameter is positive and significant.
Equations (04) and (Q40 yield the somewhat inexplicable
result that the effect of lagged income is negative. This
could be rationalized with a hypothesis which stated that,
initially, the demand for money over reacts to change in
income. Consequently, a change in the opposite direction
is required in the following period.

Both the Treasury Bill rate and the rate on long
term government bonds yield acceptable and significant re-
sults in the yearly and permanent income models. Comparison
of the elasticities of the long and short rates,presented
in Table 15, seem to indicate that the numerical values of
the long rate are significantly larger than those of the
short rate. Similar results have led other investigators
to conclude that the "demand for money in Canada is more
sensitive to changes in the long rate than in the short
rate."3 Strictly speaking, this conclusion is a proper
interpretation of the estimated elasticities. However, it

is not correct to go one step further and infer that the

 

3c, S. Laumas and P. Formuzis, "The Demand for
Money," Canadian Journal of Economics, I (August, 1968),
69A— 0

 

 

'r Wm

 

r I’.‘ ‘- #1.» 339'. "' 3‘. wt ah...

w“

80
TABLE 14

DEMAND FOR MONEY: LINEAR SPECIFICATION a

 

 

Quarterly Model
(01) * Qu = 1687.11 + .9899 Y + 75.391 r - 2066.18 D1
(75.54) (.1014) (62.366)S (218.20)
.0005 .0005 .231 .0005
1478.18 02 - 691.17 03
(166.57) (104.72)
.0005 .0005
R2 . .6307 D.W. = 1.89 rho : .9
(010 8:08 = 1371.01 + .7678 Y + 6.825 r - 1762.81 01
(69.10) (.0928) (57.050? (199.61)
.0005 .0005 .905 .0005
1262.35 02 _ 776.45 D3
(152.37) (95.81)
.0005 .0005
R2 . .5983 D. w. = 2.18 rho = .9
{02) *Qu/P: 1594.08 + .6908 Y/P +~55.502 rs — 1531.77 01
(76.59) (.1186) (56.306) (257.05)
.0005 .0005 .328 .0005
108401.40 D2 " 597068 D3
(189.85) (102.72)
.0005 00005
R2 a .4261 D.W. a 1.78 rho = .9
(Q2?) *Qa/P2 1332.66 + .6349 Y/P — 11.055 rs - 1542.0901
(70.08) (.1084) (51.520) (235.20)
.0005 .0005 .831 .0005
1096.84 02 _ 746.15 D3
(173.72) (93.98)
.0005 .0005
R2 = .5005 D.W. = 2.23 rho : .9
:0; )zsofi = .2579 AYt + .0683 AYt'l + 49.2) Arg — 119.93A r
(.1124) (.1082) (43.92) (43.94)
.025 .529 .266 .008

t-l
s

 

 

81

TABLE l4—-continued.

 

I 620 .9 D1 " 305-7 D2 - 31.5.5 D3 + 465.0 D),
2 C

( 36 ~) (392.9) (228.0) (152.9)
.010 .294 .171 .003
R2 = .3299 D.W. = 2.48
(Q3')A Q: = .3057 AYt +..11651‘vt-1 - 1.089 Ar; — 115.54 Arg-l
(.1267) (.1219) (49.520) (49.54)
.018 .343 .983 .023
— 835.7 Dl - 491.3 D2 - 409.3 03 + 507.7 DA
(266.7) (262.6) (257.1) (172.5)
.003 .066 .116 .004
R2 : .4602 0.w. = 2.54
(Q4) %Q3 = .0605 + .0468 %Yt — .0637 %Yt‘l +-.0117 8r;
(.0081) (.0319) (.0309) (.0088)
.0005 .147 .043 .191
_ .0232 firg-l — .0307 D1 — .0378 D2 — .0494 D3
(.0089) (.0306) (.0120) (.0123)
.011 .316 .011 .001
R2 = .4085 D.w. : 2.41
(04') .0: z .0706 4 .1313 %Yt — .0628 815’1-+ .0099 8r:
(.0130) ( 0514) (.0497) (.0143)
.0005 .013 .211 .488
_.0310 flrg-J— .0790 01 — 10737 02 - .0832 D3
( 0144) (.0200) (.0194) (.0197)
.034 .0005 .0005 .0005
R2 z .4776 D.W. : 2.33

Yearly Model

(Y1) Mu : 2771.04 + .5817 Y — 654.69 r1
(699.05) (.0339) (334.69)
.001 .0005 '062
R2 = .981 D.W. = .406

(Yl') M8 = 2702.78 + .3357 Y — 485.15 r1
(514.78) (.0249) (245.92)
.0005 .0005 .061
R2 z .975 D.w. = .464

 

...-1.x - ' ~ ' 1- " r drum" Fizz-1‘ - . _._.. - 5 me: (

82

TABLE l4--continued

 

(Y2) * Mu = 548.53-+ .3424 Y - 64.75 rs

(154.53) (.0452) (111.45)
.002 .0005 .567

R2 = .737 D.w. = 1.12 rho = .9
(Y2') Ma : 1283.31-+ .3897 Y - 700.06 rs
(360.68) (.0347) (229.49)

.002 .0005 .006
R2 : .952 D.W. : 1.88
(Y3) Mu/P 2 3499.2 + .5363 Y/P — 895.83 r1
(764.2) (.0841) (481.79)
.0005 .0005 .076
R2 = .863 D.W. = .3997
(13') Ma/P : 4240.2.+ .3153 Y/P - 684.70 r1
(583.0) (.0642) (367.57)
.0005 .0005 .075
R2 = .753 D.W. a .419

(Y4) 1 Mu/P = 991.18 + .1954 Y/P — 51.405 r

(314.18) (.1026) (160.777) 8

.004 .083 .753
82 e .125 D.W. = 1.11 rho : .9
(14') Ma/P : 1457.88 +—.4021 Y/P - 822.51 rs
(1054.78) (.0675) (264°89)
.176 .0005 .005
R2 = .729 D.W. : 1.73

Permanent Income Model

 

(p1) * Mu : 3620.4 + .4457 Yp — 52.109 rs
(1341.2) (.0433) (86.353)

.011 .0005 .552
R2 = .836 0.w. = 1.35 rho = .9
(Pl') Ma 2 1349.2-+-.4315 Yp — 752.65 rs
(357.1) (.0390) (236.68)
.001 .0005 .004

R2 : .951 0.w. : 2.047

 

83

TABLE 14——continued.
(P2) * Mu =(4l27.4 +.4996)Yp — 486.30 r1

 

152.9) (.0425 (182.44)
.002 .0005 .014
R2 = .871 D.w. = 1.379 rho : .9

(P2') M8 = 3567.2 +—.3539 Yp - 791.22 r1

(817.8) (.0390) (366.29)
.0005 .0005 .041
R2 = .941 D.W. = 1.44
(P3) * Mu/P = 3349.6 +-.4327 yp - 127.71 rs
(1381.6) (.0833) (147.22)

.088 .0005 .394
R2 z .5715 D.W. = 1.33 rho = .8
(P3') Ma/P_ 1471. O6-+ .4314y — 850.59 r
a (8650) ( 0598) VP (224.78) S
.10 2 .000 5 .001
R2 = .788 D.W. = 2.18
(P4) * M u/P = 3590 l-+ 5575y - 853. 80 r1
(1533 0) (. 0831)yp (292 48)
..028 .008
R2 = .674 D.W. a 1.33 rho = .8

(P4') Ma/P .- 4595.7-+-.3614 yp - 883. 50 r1
(678. 9) (.0616) (358. 40)
005 .0005 .021

R2 = .731 D.w. = 1.41

 

SYMBOLS
Quarterly Model
Qu — Non—seasonally adjusted quarterly Net Output.

Qa - Nonseasonally adjusted quarterly Net Money.

P - Quarterly implicit price deflator.

Y - Quarterly Net National Product, seasonally adjusted.
rs — End of quarter paid on three month Treasury Bills.

D- - Seasonal dummies.

 

81+

TABLE 14--continued.

 

* - Equation fitted using first order autoregressive scheme

rho- Value of first order autocorrelation coefficient.

Yearly and Permanent Income Models.

Mu
Ma
Y

Yp

yp
r3
r1
P
*

Annual average Net Output.

Annual average Net Money.

Implicit price deflator.

Net National Product.

Permanent income, in nominal terms.

Permanent real income.

Annual average rate paid on three month Treasury Bills.
Annual average rate paid on long term Dominion Bonds.

Permanent price level.

Equation fitted using first order autoregressive scheme.

rho- Value of first order autocorrelation coeffecient.

 

 

85

long rate is, in some sense, "more important" than the short
one. Theorv implies, and inspection of the Canadian data
confirms, that fluctuations in the short rate are larger
than those in the long rate.h Thus, when a given stock of
money is regressed against these two rates, one would expect
the estimated coefficient on the long rate to be larger than
that of the short rate.

Another author uses the higher R2, obtained in his
results, of the short rate to justify its choice as correct.5
The use of this technique on the evidence presented in this
study has, at least, doubtful validity. The failure of a
first order autogressive scheme to yield acceptable Durban-
Watson values in many of the equations may indicate the
presence of higher order autocorrelation or other serious
specification errors. It would, of course, be nice to
point to one rate as truly representing the alternative
costs of holding money. But, until better criteria are de-
veloped, the choice of a "correct" rate, if it exists,

must be postponed.

C. Incgme and Prices
Other empirical studies have used current income,

non-human wealth, and permanent income, assumed to be a proxy

 

AB. A. Musgrave, The Theory of Public Finance, (New
York: McGraw Hill Book Company, 1959), p. 596.

5David Laidler, The Rate of Interest, p. 547.

 

 

 

 

86

for total wealth, as scale variables in the demand for money.
Unfortunately, a series of non-human wealth is not available,
to the best of the author's knowledge, for Canada. However,
values for permanent income were computed using the vearly
values of Net National Product and a series of expotentially
declining weights.6 Thus, two of the three commonly used
scale variabms employed by others are represented in the
money demand estimates.

The analysis of Chapter II was constructed under the
assumption that monev, just as all other capital goods, was
demanded for the services which it rendered. The amount of
money in a wealth portfolio is usually assumed to be directly
related to the size of the portfolio. If wealth is zero,
the demand for money will also be zero. Thus, combined with
one of the usual linearizing assumptions needed for empirical
testing, this analysis leads to the often investigated
hypothesis that the demand for money has unit elasticity

., ’7
with respect to wealth.“

 

6These weights were kindly provided by G. L. Laumas
and P. Formuzis. It will be assumed that the same adjustment
process is appropriate for real income and the pricc level.
Thus, permanent prices and permanent real income are calcu-
lated using the same weights as permanent income.

7Most studies assume, as is done here, that the cor-
rect specification is linear in the untransformed variables
or in logs. However, Zaremka has demonstrated that these two
specifications are Simply limiting cases of the general formo
See Paul Zaremka, "Functional Form of the Demand for Money,"
Journal of the American Statistical Association, 63 (June,
L968), BOZmEIl.

 

87

Alternative analysis, which seems to imply a differ;

ent hypothesis, states that the demand for money is directly
related to transactions, represented by current income. Since
the purpose of money is to facilitate transactions, the demand
for money may be expected to be positively related to the
number, or value, of transactions. If transactions were

zero, the demand for money, because there is then no use

for this type of good, would also be zero. Thus, this analy—
sis also implies a unit elasticity with respect to this

scale variable.

These two hypothesis, despite their use of different
scale variables, are not in confliCt. An increase of items
in a wealth portfolio can be obtained, in a money economy,
only by undertaking transactions. Thus, it is quite use-
less to argue whether the increase in wealth or transactions
is the cause of an increase in the demand for money.

Equations (Y5) through (Y8) and (P5) to (P8)in Table 1%
seem to indicate an estimated elasticity of close to inity
for the demand for money with respect to current yearly and
permanent income. However, the quarterly elasticities,
presented in (Q5) and (Q6) are much lower than the corres—
ponding yearly and permanent income estimates.

The quarterly, yearly, and permanent income elastici-
ties of the demand for money display a characteristic found
in the demand for many other goods: namely, the value of the

elasticity increases as the time period represented by the

 

88

data points is lengthened. Thus, it may be more appropriate

to view the income variables in light of an expectation
process rather than the simple wealth or transactions hy-
potheses presented above. A very simple version of this
hypothesis would state that permanent income is a "long run"
estimate of "expected" income, or the return on wealth, in
that it allows the incorporation of, properly weighted,
realized incomes of many past periods. Yearly and quarterly
income are shorter run estimates because they incorporate
progressively fewer periods.

If an expectation process of the above type is
assumed, then one may expect that the income elasticity of
the demand for money would approach unity as "expected"
income is approximated. In other words, quarterly, yearly,
and permanent income, being progressively better estimates
of "expected" income, would yield elasticities closer and
closer to one.

As was previously stated, the quarterly estimates of
the income elasticities are substantially below those of the
yearly and permanent income models. However, the estimates
of the current yearly and permanent elasticities are all
quite close to unity. Thus, one is unable to distinguish be-
tween the effect of yearly and permanent income. In other
words, if the long run elasticity is one, this evidence seems

to indicate that the entire adjustment is completed

 

 

)-.- .IKNK“

 

89
TABLE 15

DEMAND FOR MONEY: LOG SPECIFICATION a '

 

Quarterly Model

 

(05) * log(Qu) :.3232 +-.29OO log(Y)-+ .0367 log(rs)

(.0158) (.0507) (.0165)
.0005 .0005 .029
-00459 01 _ .0344 D2 — .0171 03
(.0084) (.0066) (.0042)
.0005 .0005 .0005
R2 3 .43242 D.W. = 1.19 rho = .9
(Q5') * log (Qa) = .2938-+ .3593 log (Y) +-.0245 log (rs)
(.0181) (.0582) (.0189)
.0005 .0005 .199
- .0635 01 — .0316 02 — .0463 D3
(.0097) (.0048) (.0075)
.0005 .0005 .0005
R2 = .5069 D.w. - 1.69 rho = .9
(Q6) === log (Qu/P) .-. .3552—+ .1837 log (Y/P) + .0263 log (rs)
(.0152) (.0491) (.0142)
.0005 .0005 .069
1 .0303 01 _ 00223 02 _ .0144 D3
(.0080) (.0032) (.0036)
.0005 .001 .0005
R2 z .3051 D.W. : 1.42 rho = .9
(Qé') * log(Qa/P) :: .3207 + .2899 108(Y/p) + .0147 log (rs)
(.0187) (.0605) (.0175)
.0005 .0005 .403

- o0504 D1 — 10372 D2 - .0294 D3

(.0099) (.0076) (.0094)
.0005 .0005 .0005

R2 = oAZOLp Do‘No B 1090 I‘hO I 09

(Q7) * log(Qu) 3 «3609 é‘oléhb lOg(Y)‘+-l.O869 log (P)

(.0158) (.0513) (.2246)
00005 .002 .0005

 

 

 

 

90

TABLE 15--continued.

 

* .0216 log(rs) - .0275 D1 — .0199 D2 - .0141 D3

 

(.0147) (.0083) (.0064) (.0036)
.147 .001 .0005 .0005
R2 = .5766 D.W. : 1.536 rho : .9
(07') * 108(Qa) 3 ~3232 +-.2617 log(Y) 1..8453 log(P)
( .0197) (.0639) (.2797)
.0005 .0005 .003
8 .0127 log(rs) — .0492 D1 - .0367 02 — .0292 D3
(.0183) (.0103) (.0080) (.0046)
.489 .0005 .0005 .0005
R2 z .5631 D.W. - 1.942 rho = .9
Yearly Model
(Y5) log(Mu) = «.4411 y-l.1058 log(Y) - .4813 log(rl)
(.2404) (.0726) (.1463)
.079 .0005 .003
R2 . .966 D.W. = .467
(YS') 10g(Ma) = .0865 + .9349 lOg(Y) — .3741 log(r1)
(.2359) (.0713) (.1436)
.717 .0005 .016
R2 = .957 D.W. = .445
(Y6) logiMu) : - .8520 4 1.1432 log(Y) — .1540 log(rs)
(.2888) (.0692) (.0413)
.007 .0005 .001
R2 z .974 D.w. .749
(Y6') log(Ma) : - .8685 + 1.1175 log(Y) - .2147 log(rs)
( 2925) (.0700) (.0418)
.007 .0005 .0005
R2 . .966 D.W. g 2.008
(Y7) log(Mu/P)=~' -,.5473 81.1186 iog(Y/P) - .4043 log(r1)
(.4637) (.1217) (.1338)
.250 .0005 .006
R2 8 .873 D.w. = .518

 

91

TABLE 15--continued.
(Y7') log(Ma/P)= .1089 + .9448 log(Y/P) - .4698 log(rl)

 

 

(.5467) (.1434) (.1577)
. .844 .0005 .007
R2 z .7346 D.w. : 1.207
(Y8) log(Mu/P)= — .8835 + 1.1467 log(Y/P) - .1157 log(rs)
(.5801) (.1351) (.0403)
.141 00005 .008
R2 z .869 D.W. = .599
(y8') log(Ma/P)= -.9015 + 1.1220 log(Y/P)- .1838 log(rs)
(05759) (.1342) (.0399)
.131 .0005 .0005
R2 = .809 D.w. = 1.786
(Y9) log(Mu) = .4844 + .8399 log(Y)‘+-6035 log(P)
(.7223) (.1652) (.3020)
.509 .0005 .058
_ .1571 log (rs)
(.0389)
.001
R2 = .9776 D.w. : .799
(Y9') log(Ma)= .2021 + .8748 log(Y) + .4830 log(P)
(.7548) (.1727) (.3157)
.791 .0005 .140
- .2171 log(rs)
(.0407)
.0005
R2 1 .9689 D.W. : 2.139
Permanent Income Model
(P5) log(Mu) = w .5545 + loO855 log(Yp) - .1455 log(rs)
(.1583) (.0384) (.0239)
.002 .0005 .0005
R2 z .990 D.w. : 1.36
(pS') log(M )= - .4950 ~ 1.0412 log(Y ) - .1953 log(r )
a (.2415) (.0585) p (.0365) S
.057 00005 .0005

R2 g .972 D.w. : 2.332

 

 

92
TABLE 15--continued.
(P6) log(Mu) = .0344 + 1.0056 log(Y ) - .3591 log (r1)

(.1181) (.0366) p (.0846)
.773 .0005 .0005
R2 = .986 D.w. : .349
(P6') log(M ) = .2851 +.9024 log(Y ) - .3924 log(r )
a (.1911)(.0593 p (.1369) 1
.149 .0005 .009
R2 z .954 D.w. = 1.271

(P7) * log(Mu/P): -.O426 + 1.0873 log(yp) -.3294 log(rl)

(.0562) (.1392) (.1006)
.453 .0005 .003
R2 = .752 D.W. : 1.216 rho = .7
(137') log(Ma/P): --5796 + 08285 log(Yp) 7-368h log(rl)
(.4072) (.1076) (.1262)
.167 .0005 .008
R2 : .961 D.W. : 1.069
(P8) log(M /P)= -.5946 + 1.0873 log(y ) — .1092 log(r )
u (.2696) (.0633) P (.0201) S
.037 .0005 .0005
R2 = .961 D.W. : 1.069
(P8!) log(Ma/P)= -.3936 + 1.0112 log(yp)- .1632 log(rs)
(.3991) (.0937) (.0296)
.334 .0005 .0005
R2 = .872 D.W. = 2.376
(P9) log(Mu) = -.1052 f .9845 log(Yp) +.2328 log(p)-JU%4kg&g)
(.5280) (.1196) (.2609) (.0270)
.844 .0005 .381 .0005
R2 = .991 D.w. = 1.445
(P9’) log(M )= «”0632 +.944l log( H.223? log(p)-.20581c>g;(r )
a (.8135)(.1843) Ypmom) (.0416) S
.939 .0005 .583 .0005
R2 = .972 D.w. = 2.326

 

aSymbols defined in Table l.

 

 

 

93

within one year.8

It has been asserted that the approximate equality
of the coefficients on the real and nominal scale variables
implies that the demand for money is homogeneous one in
prices.9 Similar comparisons for the results of this study,
1. e. comparing (Y6) and (Y6') with (Y8) and (Y8') could
lead to the same conclusions.

Rather than relying on indirect inference, the price
level was explicitly included in the quarterly, yearly, and
permanent income models. The results are presented in (Q7),
(Y9), and (P9)° A disturbing phenomenon is observed when
the estimated price elasticities in the three models are
compared. The price elasticities are close to one in the
quarterly formulation, the yearly elasticities are about
.55, and the permanent price elasticifies are not signifi-
cantly different from zero. If these estimates are accurate,
they imply that changes in prices have no long run effect
on the demand for nominal balances. This is certainly con-
trary to the accepted notion of the reaction of the demand
for money to price level changes.

An explanation which would rationalize these

 

8An alternative hypothesis stating that the differ-
ence between the effect of current yearly and permanent in-
come is not accurately measured with this specification
is presented below.

9Allen H. Meltzer, "The Demand for Money: The Evi-
dence from Time Series," Journal of Political Economy,
LXXI (June, 1963), 225.

 

94

elasticities is obtained by assuming that the "true" money

demand specification is

(111-1) 10g (M/P) = a + b 105:, (Y/P) — c 16;; (r).

.This specification may than be rewritten as

(III-2) 10g (M) = a +b 16g (Y) + (l-b) 16g (P) - c 16.2; (r).
Thus, if (III—l) is the correct specification of the model,
the coefficient on log (P) in (III-2) is really an inde-
pendent estimate of the real income elasticity.

In the quarterly case, the coefficients of log
(P) in (Q7) and (Q7') are, taking account of the standard
errors, approximately equal to one minus the estimated co-
efficients on log (Y). These indirect estimates are also
not inconsistent with the direct estimates of the real income
elasticities presented in (Q9 and (Q5'). Likewise, all of
the estimates of the permanent real income elasticities are
not far from the predicted value of unity. However, in the
yearly current income model, the coefficients on log (P)
predict real income~elasticit1es of about .55 while the co-
efficients on log (Y) in (Y7) and (Y8) are not significantly
different from unity.

The two sets of estimates of the elasticity of the
demand for money with respect to current yearly income are
in conflict. The direct estimates indicate an elasticity of
close to unity, which is not significantly different from
' the permanent income estimates. On the other hand, the in—

direct estimate of the current real income elasticity, just

 

95

presented, is about one half that of the direct estimate.

Consecutive values of yearly current income are
highly correlated with each other and with permanent income.
Thus, current income may be a proxy for permanent income
which results in estimates of permanent, rather than cur—
rent, income elasticities. In an attempt to obtain estimates
free from this serial correlation, the relevant series were
converted to percentage change. Then, the following form was
run using the transformed data of the quarterly, yearly,
and permanent income models,

(III—3) %M = b O/OY.

Where 96 represents percentage change and Y is current or
permanent income. Using this specification, b would be a
direct estimate of the income elasticity.

The estimated elasticities for the quarterly data
are presented in Table 3, equations (Q8) and (Q9). They are
lower than, but of approximately the same magnitude as, the
estimates obtained from the various log specifications.
Also, they are consistent with the estimates obtained from
the coefficients on log (P) in Table 15, equations (Q6)
and (Q6'). Thus, all the estimates seem to imply that, even
in a period as short as a quarter, the demand for money
begins to adjust to changes in income.

The results, in both real and nominal terms, for
current yearly and permanent income are also presented in

(YlO), (Yll), (P10), and (P11) in Table 16. The estimated

 

 

 

96
TABLE 16

DEMAND FOR MONEY: PERCENTAGE CHANGE ELASTICITY ESTIMATESa

 

(Q8) %Qu = .0669 %Y - .0001 D1 + .0100 D2 + .0022 D

(.0328) (.0081) (.0054) (.0046)
.045 .988 .072 .634

+ .0329 D
0 1) 4

(.O 6
.0005
R2 = .2841b D.W. = 2.269
(08!) %Qa = .1505 %Y - .0220 D1 + .0009 02 — .0061 03
(.0508) (.0127) (.0071 (.0085)
.004 .086 .918 .396
+ .0592
(.0094)
.0005
R2 = .4242b D.W. : 2.245
(09) %Q /P): .0737 %(Y/P) — .0102 D 4 .0019 D — .0047 D
u (.0434) (.0102) l (.0068) 2 (.0060) 3
-O93 «325 .783 .440
+ .0251 D
(.0081) h
.003
R2 : .2002b D.w. : 1.886
(Q9') %(0 /P)= .1684 %(Y/P) - .0334 D1 — .0077 D -.0113 D
a (.0599) (.0142) (.0094 2 (.0084) 3
.021 .410 .181
+ .0536 D4
( .0112)
.0005
R2 = .3647b D.W. = 2.084

Yearly Model
Y10) %M = .5865 %Y
( u (.1106)

.0005
R2 = — .7197b D.W. : .6311

 

97

TABLE l6--continued.

 

(YlO') %Ma = .7614 %Y

 

(.2093)
.001
R2 : .061b D.W. : 2.814
(Yll) %(Mu/P) = .4286 %(Y/P)
(.1590)
.012
R2 g - .1665b D.W. = .9673
(Yii!) %(Ma/P) = .5700 %(Y/P)
(.2876)
.058
R2 = .014b D.W. = 2.737
_ Permanent Income Model
(P10) %Mu : .8160 %Y
(.0912) P
00005
R2 = .1223b D.W. = .8503
(P10') %Ma =..9195 %Yp
(.2357)
.001
R2 : .1060b D.w. = 2.941
(P11) %(Mu/P)’ — .7519.%(yp)
( 31549)
.0005
R2 2 -216b D.w. = 1.216
(Pll') %(Ma/P): .8356 %(yp)
(.3266)
.015
R2 : .101b D.w. : 2.906

 

a. Symbols defined in Table l.

b. When the constant is restricted to zero, R2 is defined as
unity minus the Sum of Squares Regression divided by the
Sum of Squares of the dependent variable about its mean.
Thus, it may be negative and does not reflect the signifi—
cance of the estimated relation.

 

98

yearly elasticities are all significantly lower than the

direct log estimates of Table 15, equations (Y5) through
(Y8), and different from one. The permanent income elasti—
cities for the Net Output series are also significantly
different from the log estimates and from unity. However,
the permanent income elasticities of the Net Money series
are quite close to the log estimates and are not different
from unity.

Conveniently, the elasticities estimated in this
way may be arranged in ascending order as one moves from the
short to the long run. The quarterly income elasticities
are the lowest, the yearly are larger and significantly lower
than one, and the permanent income elasticities are close
to unity. These results would not contradict a hypothesis
which stated that the long run income elasticity of the de-
mand for money is unity and that the estimated values tend
toward this limit as one extends the time period allowed
for adjustment. The period needed to adjust money holdings
to a change in income is certainly longer than a quarter
and probably greater than a year.

The estimated permanent income elasticities of Net
Output and Net Money present a rather puzzling situation.
The Net Output elasticities, in (P10) and (P11) are not
statistically different from the corresponding Net)Money
estimates, but they are significantly less than unity. One

may expect that the income elasticity of debt is larger than

 

 

 

 

99

that of money. This would tend to increase the elasticity
of Net Output over that of Net Money.lo Thus, these results
are somewhat inconsistent with the hypothesis concerning the

debt component in Net Output.ll

D. Adjustment Processes in the Demand for Money

In order to more carefully examine the adjustment
process implied in the various demand functions, a variant
of a stock adjustment model was estimated. It is assumed
that the long run "desired" stock of money is a function of
"expected" income and interest rates.
(III-4) Md = b0 + b1 Ye — b2re.
The superscripts "e" and "d" denote "expected" and "desired"
magnitudes. The observable stock of money is related to the
"desired" by means of an adjustment mechanism.
(111-5) Mt . Mt_l + s(N§ - thl).
Mt is the observed stock of money and "8" represents the
amount of the divergence from the "desired" made up in one
period. When the variables are in log form, "5" is the

elasticity of adjustment. Substituting into (III-4) for

Md, we obtain,

 

10For example, Friedman's well known conclusion that
M2 is a luxury good.

11The author has estimated that the elasticity of
Trust and Loan Company Debentures, a pure debt item similar
to Personal Savings Deposits, with respect to current yearly
income is about 1.15. If the permanent income elasticity
for the debt component of money-debt is similar to that of

 

 

 

 

100
(III-6) Mt erbO + (l-s)Mt_1 + sbl Ye - sbzre.

Unfortunately, the inclusion of "expected" interest rates and
income imply similar adjustment mechanisms for these vari-
ables. When these mechanisms are substituted into (III-6),
these are not sufficient conditions to determine all par-
ameters.12 Thus, it is necessary to choose values for "ex-
pected" interest rates and income such that they already
reflect the adjustment process.

Evidence from the United States seems to show that
the adjustment of interest rates is completed in one yeargh
Thus, it would be reasonable to include the current value of
this variable. Previous analysis has implied that the
elasticity of the demand for money with respect to wealth,
or long run "expected" income is unity. Reversing the
logic, one may say that the unitary elasticity, found in

nearly all the specifications, implies that permanent income

 

Debentures, the "predicted" elasticity of Net Output might
not be too much larger than that of Net Money.

12Edgar Feige, "Expectations.and Adjustments in the
Monetary Sector," American Economic Review, XVII (May,

1967), 462-473.

13For example, one study did not consider the ad—
justment in income and, using the current value of this
variable, obtained a quarterly estimate of .32 for "s". If
there actually exists an adjustment in income, this estimate
is much too high. See R. Teigen, "Demand and Supply Functions
for Money in the United States," Econometrics, XXXII (October,
1961;) , [588-489 0

ll’zFeige, p. 470.

 

 

 

101

has taken account of all the adjustments in this variable.

Therefore, using current interest and permanent income, one
is left with only the lag in money balances. Estimation of
(III-6) will then lead to a direct estimate of the rate of

adjustment in money.

The estimates of (III-6) are presented in Table 17
equations (P12) and (P13). In both log and non-log form,
the adjustment coefficient of Net Output is about .3 and
that of Net Money .65. These estimates seem to imply that,
in each period, twice as much of the divergence is made up
in the Net Money as in the Net Output series. However, in
neither case, is the entire discrepancy eliminated in one
period. For the United States, it was found that the ad—
justment coefficient of M2, which contains time deposits,

a debt item, was below that of Ml’ But neither was
statistically different from unity.15 Thus, the Canadian
results would not be inconsistent with a hypothesis stating
that the money holdings in a wealth portfolio are brought

into equilibrium faster than debt items.16

 

l5Feige, p. 470.

16However, these results are also not inconsistent
with a hypothesis stating that consumers are slower in ad-
justing their discrepancies of money holdings than are
firms. The Net Output series includes Personal Savings
Deposits, which are held mostly by individuals, at their
nominal value. The Net Money series give Personal Savings

 

 

 

TABLE 17

DEMAND FOR MONEY: STOCK ADJUSTMENT SPECIFICATIONa

 

Permanent Income Model

(P12) M3 : 1264.95 + .7335 M3'1 .1796Yt - 312.11 r1
(384.16) (.0878) (.0504) (129.09)

.001 .0005 .002 .024
R2 = .998 D. w. = 1. 235b
(P12!) Mg 2 2520.86+ .3023 Mt- -1 + .2772 Y; -536. 26r
(1057.76)L 2009)a (. 08662) (395 17)E
026 .146 .004188
a? = .9465 D w = 1 235b

(P13) log(Mt)=. 1601+ .7118 log(Mt'l)+ .2515 log(Yg)
(. O610)(. 0796) (0 M6 )
.015 .0005 .008

_ .0795 16g(r E)

(. 0514)
«135
R2 = .997 D w. 2 .859b
(P13!) log(ME) = .3005 4 .3814 iog(Mt 1) + .5261 log(Yp)
(.1801) (.1188 7) (.1994)
.109 .055 .013
— .2029 log(rE)
(.1594)
.216
2‘: °961 D.w. : 2.184b

 

a. Symbols defined in Table 1.

b. The Durban~Watson Statistic has doubtful validity when a
lagged value of the dependent variable is used. See Marc
Nerlove, and Kenneth Wallis, "Use of Durban-Watson Statis—
tic in Inappropriate Situations," Econometrics, Vol. 34,
No. 1, pp. 235-238.

102

 

 

 

103

The differences between the speed of adjustment of
Net Output and Net Money is consistent with the percentage
change estimates of the income elasticities. In the quarterly
and yearly estimates, the elasticity of the Net Output series
is below, although sometimes insignificantly so, that of the
Net Money series. The permanent income estimates of the two
elasticities are quite close. Thus, the two formulations
do not contradict each other in implying that the Net Money
series adjusts faster to a change in income than the Net

Output Series.

 

Deposits a smaller weight. See Harold Shapiro, "Distributed
Lags, Interest Rate Expectations, and the Impact of Monetary
Policy: An Econometric Analysis of a Canadian Experience,"
American Economic Review, LVII (May, 1967), 474.

CHAPTER IV

SUMMARY

I. Introduction

 

Chapter I presented the two major obstacles to the
construction of a money supply series for Canada: 1. The
large proportion of checkable deposits bearing explicit
interest. 2. The data published by the Bank of Canada
does not reflect the change in the post war Canadian money
producing industry. This chapter will attempt to summarize
the methods used to overcome these difficulties and the

statistical findings of Chapter III.

II. Interest Bearing and Checkable Deposits

A large portion of the output of private money
producers consists of interest bearing and checkable de-
posits. The checkability of these deposits indicates that
they are money. However, the presence of explicit interest
payments, characteristic of debt items, creates uncertainty
as to their exact position in the money stock. Consequently,
a general method for the analysis of money-debt was pre-
sented.

The marginal rates of return on each item in a

wealth portfolio will, in equilibrium, be equal. Moneywdebt

th

 

 

lOS

earns one stream of income from its use as money and another

from the explicit interest payments. Thus, the sum of these

two rates must, in equilibrium, be equal to the market inter-
est rate.

Capitalizing the two streams of income accruing to
money—debt, one has the portion of its value that is due to
it serving as money and the portion due to its being an
interest bearing debt. The first component is then included
in the money stock. Empirically, the pr0portion of the value
of money-debt that is not money can be estimated by the ratio
of the explicit rate paid on these deposits to the market
interest rate. Subtracting this portion from the value of

moneywdebt, one is left with the money component.

III. Construction of the Money Series

 

Derivation of the money stock from the data of many
different types of producers required the construction of a
general methodology for the classification of their outputs.
It was concluded that the only rationale for the exclusion
of the money held by any firm is that it be used as in
input in the production of money. Two major results folm
lowed from this analysis.

The estimate of the money production of private
money producers was biased downward. Private money producers

also manufacture the type of pure debt which requires the use

of money as an input. Balances used as "reserves" against

 

106

debt production, because they are not an input for money,
should not be excluded from the money stock. However, data
limitations did not allow the division of money holdings of
these firms into ”reserves" against money and non-money
output. Consequently, the net money production of these
money producers was calculated by subtracting their entire
money holdings from their gross output of money. The degree
of underestimation, because the "reserves" held against both
money and non-money production are subtracted from gross
output, of the net money production of different institutions
will be related to the prOportion of money in their output
of all goods. The firms which had the lowest proportion of
money in their mix of products were the non-chartered banks.
Consequently, their cmitribution to the money stock was
greatly underestimated.

The second implication of this analysis is that the
deposits of the various governmental units in the money pro-
ducing industry should certainly-be included in the money
stock. In the present setting, they are not viewed as
"reserves" against government money production and should,
therefore, not be excluded.

The empirical counterpart to the market interest
rate was taken to be a weighted average of-the rates paid on
a number of consumer debt items. Canadian moneywdebt is
sold primarily to individuals, with its rate reflecting the

risk premium prevalent in this market. .Thus, the appropriate

 

 

 

lO7

comparison rate is on items with the same degree of risk.

Each money producer was examined in detail and his
net money production was estimated. First, the debt portion
of his money-debt was estimated and eliminated. Then the
remainder of the value of his output of money-debt was added
to his output of pure money and "reserves" were subtracted.
The result was designated "Net Money." Another entry was
calculated by foDowing the same procedure but not eliminating
the debt portion of money-debt. This series was titled
"Net Output."

By summing across the money producers, two estimates
of the money stock were obtained. The first, the sum of the
Net Money of each producer, corresponds to the view that
interest bearing and checkable deposits consist of a money
and a non-money portion and only the former should be inn
cluded in the money stock. The second, the sum of the Net
Output, may represent a more traditional view of the
money stock.

Identical output data was not available from each
type of producer for every time period. Consequently, a
number of series, differing in coverage and time interval,
were constructed. The basic quarterly series runs from
lst quarter 1939 to tth quarter 1966 but excludes a number
of important producers. Complete coverage is obtained in a
yearly series encompassing the same period.

The comparison rate, used to estimate the debt

 

 

 

108

portion of money-debt, was calculated only as an annual
average. Consequently, the movement of Net Money within
each year can be expected to be identical to that of Net
Output. Simple graphical comparisons of the two quarterly
series confirmed the suspicion that there is little differ-

ence between their movements over time.

IV. The Demand for Money in Canada

The demand for money was chosen as a topic which
could present the attributes of the constructed series. How-
ever, it was clearly pointed out that available hypotheses
were not sufficient to test the analysis underlying their
construction. Specifically, if the analysis is correct, then
the only difference between Net Mcney and Net Output is the
debt component of money-debt, included in the latter. But,
without further information concerning the demand for debt,
very little may be implied from the money demand estimates
concerning the validity of the exclusion of the debt
component.

Various specifications of the demand for money were
run using quarterly, yearly, and permanent income models.
Results for Net Money and Net Output were presented side by
side.

In the quarterly model, current income yielded
significant coefficients with the correct sign.However, all
interest rate variables were either insignificant or perverse

in their effect. Only in a first difference model,vdth the

 

 

lO9

interest rate lagged one period, were the results consistent
with a priori expectations.

Current yearly income, permanent income, and rates
on long and short term government bonds yielded significant
coefficients with the correct sign. However, the log esti—
mates of the elasticity of the demand for money with respect
to both current yearly and permanent income were close to
unity. A percentage change specification was used to dis-
tinguish between the effects of these two concepts of income.
This formulation yielded a permanent income elasticity of
close to unity but a much lower current income elasticity.

In most models and specifications, the estimate of
the income elasticity of Net Money was larger, although some-
times insignificantly so, than that of Net Output. In order
to explicitly investigate these implications, a stock ad~
justment model was estimated. The results indicated that
the speed of adjustment of Net Money is almost twice that
of Net Output.

In summary, simple statistical techniques seem to
indicate that there are differences, at least on the annual
level, between the properties of Net Money and the Net Output
series. However, available hypotheses are not sufficient to
test if these differences are consistent with the analysis

underlying the construction of the two series.

 

APPENDIX I

MONEY PRODUCTION IN PRE—CONFEDERATION CANADA

I. Introduction

The cultures of the United States and Canada have
developed along parallel lines. However, certain historical
events have caused the institutions of the two countries to
differ. Thus, even though the broad outlines are similar,
one cannot generalize about the structure of specific
Canadian industries from their American counterparts. In
particular, the structure of the Canadian money producing
industry is quite different from the American money producing
industry. The differences between the current form of the
two industries is, of course, the result of influences
occuring throughout their histories. But two of the primary
characteristics of Canadian money production, the coexistence
of a regulated cartel with groups of non-regulated money
producers and the production of money-debt, may be traced to
events occurring prior to the Confederation.

The purpose of this section is to trace the de-
velopment of the Canadian money producing industry from its
inception to the time of the Confederation. The emphasis will
be on the two characteristics mentioned above. The result

110

 

111

will not be a description of the modern Canadian money pro—
ducing industry. But, h0pefully, a feeling will be given

for the historical context within which the modern insti-
tutions operate. Then, certain qualities of the products and

the data of the modern money producers will be more meaningful.

II. Commodity Money

During the 16th, 17th, and 18th Centuries, Canada
was sparsely settled frontier with only a few settlements
scattered along the eastern seaboard and the Great Lakes. The
economy was correspondingly primitive and simple.

In this primitive setting, most economic units
were self-sufficient and production for market sale was
minimal. The need for a standardized medium of exchange was
small because direct barter was capable of satisfying the
needs of trade. However, within certain geographic areas,
different commodities were at one time or another used to
eliminate the inconvenience of direct barter. Beads, blankets,
and tobacco were the usual commodities chosen, with the latter
even being elevated to the status of dominant money in several
communities.1

In the populated areas, and extending somewhat into
the frontier, assorted metal coins were used as money.

American, British, Spanish, and Latin American coins were

 

1Canada, The Canada Yearbook, 1938, p. 892.

112
used in both domestic and international trade.2 Until the
Confederation, the various provinces designated one or the

other of the coins as specie, dominant money, and set the

prices of the other coins valid for that province.3

III. Paper Money

 

The first recorded use of paper money in Canada was
in 1685. At this time a French official out playing cards into
quarters, wrote amounts on them, and used them to buy goods and
services.“ This paper money worked well and was convertible on
demand into Bills of Exchange drawn on the French government.
During the next 100 years, as the French positiOn deteriorated,
the repurchase clause was not consistently honored as sub-
sequent issues of French paper money increased the supply many
thousandfold.5 The value of this money decreased and by 1763
its price had dropped to zero and it disappeared from use.6

Expectations of capital loss, derived from the French

experience, reduced the demand for paper money produced by any

government or by a firm whose major product was money.

 

21bid.

 

3Victor Ross, The History of the Canadian Bank of
Commerce (Toronto: Oxford University Press, 1920), I, p. A.

 

thid., p. 5.

5B. E. Walker, A History of Banking in Canada (Toronto:
1920), p. 8.

 

6Ross, Bank of Commerce, pp. 3-6.

 

 

 

 

113

Consequently, during the 55 year period following the French and
Indian Wars several attempts to form banks met with a total lack
of interest.7 There was, however, production of paper money by
firms whose primary business was in other areas. This money
consisted of the notes of exchange of the most prominent busi-
nessmen. The supply of these notes was very small.8 Conse-
quently, the problems of the French with the repurchase clause
were avoided.

During the War of 1812, the British army produced
paper money in order to buy goods and services. The repurchase
clause of this money was in terms of interest bearing govern—
ment bonds.9 The army produced only limited quantities of money
and promptly honored any demand for repayment, probably by
simply printing bonds. The Canadians became convinced that
they would not suffer a capital loss by holding this money
and the old fears and distrust of paper money disappeared.

With this example of a "well managed" paper money firmly
in mind, two groups of merchants in Montreal and one in Quebec
attempted to incorporate into firms which would produce money.
At first their incorporations as joint stock companies was re-
fused by the legislature. However, the demand for paper money

was great and the profitability of money production promised

 

7Roeliff Morton Breckenridge, The Canadian Banking
System, 1817-1890 (Toronto: Printer Unknown, 189A), pa 8.

8Ross, Bank of Commerce, p. 8. 9Ibid., p. 9.

 

 

 

to be high.

114

Consequently, these groups went ahead and operated

under agreements of partnership. Finally, in 1821, the legis-

lature granted three joint stock charters to enter the busi-

ness of banking.

10

The charters granted to these banks were probably the

single most important influence on the future structure on the

Canadian money producing industry. A brief outline of one

charter, that of the Bank of Montreal, will serve as to repre—

sent the attributes of all. The Bank of Montreal may

1.

2°
3.

h.

"hold real estate only to the value of 1,000
pounds.

sue and be sued. .

issue promissory notes intended to circulate as
money and payable in gold or silver coin current
Bead dominant mone by law of the province.
receive de osits an deal in (a) bills of ex-
change, (b? discount notes of hand and promissory
notes and to receive the discount at the time of
negotiation, (c) deal in gold and silver coin and
bullion, and (d) the sale of stock pledged for
money lent but not redeemed.

to take and hold mortgages on real property for
debts contracted in the ordinary [emphasis minfl
course‘of the dealing, but on no account to lend
on land, mortgages, nor to purchase them on any
pretext except here permitted,

not to demand or to receive more than the lawful
interest of six per cent per annum on any of its
dealings.

have total liabilities not in excess of "treble
the amount of capital stock paid in, plus a sum
equal to moneys deposited with it for safe
keeping."12

 

Almost immediately two of the characteristics of

modern Canadian banking made their appearance. To counter the

 

loBreckenridge, S stem, pp. 19—21. llIbid. pp. 2h-25.
12Ibid.

 

 

 

115

entry of competitors, these three banks began to open branches
in the other population centers of the province. At the same
time, they began to cooperate closely in the clearing of
notes and checks.13 Through time, this cooperation was to
develop into a cartel with the clearing house, by restriction
of use, becoming the means for controlling members and re—
stricting entry into the money producing industry.”+
The production limit imposed on the banks, es—
sentially at their own volition because the charters were
but a copy of their own acts of incorporation might seem
illogical. One explanation might be that they believed that
their most profitable output was at some amount less that
triple their paid in capital. Then, the inclusion of a
maximum was simply propaganda. But this would not explain
the inflexibility of establishing the unchangeable legal
maximum which cannot be quickly altered if circumstances
changed. A plau51ble explanation is obtained by remember-
ing that the producers of money, the merchants, were also
the largest demanders of money. They would gain most from
having a paper money of reasonable value and be greatly
hurt if the supply of money increased to such an extent that
its price dropped to zero and could not be used. Thus, be-

cause of their gain as users of money outweighed their

 

l3Ibido, pp: 27"280

l“Joseph French Johnson, The Canadian Bankin
System (Washington: U.S. Government Printing Office,l9lO),
po 13h»

 

116

profit as producers of money, the money producers subjected

themselves to this output maximum. Another aspect which is

to play an important role later is the form of this production
limit, which was a precedent followed in later periods. It
stated that Eggal "liabilities" of the bank was not to ex-
ceed triple the paid-in capital. There was no restriction

on the composition of total "liabilities".

The position of the money producing cartel was en-
hanced by legislation of the 1830's. At this time the
banking lobby was able to obtain legislation which re-
stricted note issue to chartered banks or to institutions
which were granted licenses by the legislature.15 These
acts did not mention or restrict deposits or the granting
of checking privileges.

During the same period there was the first attempt
to legally remove the barriers of entry into the money prom
duc1ng industry.16 Several bills were proposed concerning
uniform banking legislation which would allow "free
banking". These attempts failed with the passage of an
act in 1838 which stated that it "is inconsistent with due
regard to the protection of commerce and the welfare and

security of the people, that any person or number of

 

l5Breckenridge, System, p. 35.

16Roeliff Morton Breckenridge, The History of Bank-
ing in Canada (Washington, Government Printing Office, 1911),
p. 34.

 

rt~ ‘*

‘f

 

 

 

117

persons, some of whom may be of doubtful solvency, should
be allowed, without legislative authority, to issue
promissory notes as money.17

There seem to have been a few note producers out-
side the cartel operating as banks under agreements of
partnership. Their operations were either in direct con-
flict with the law or they exported their money to the
United States. In either case, these banks were of little
consequence. By 18h1, all but one of them either went out
of business or were bought up by members of the cartel.l8

In contrast to these short lived "private banks",
several non-cartel money producers began operation which
have continued to the present. The first are the "savings
banks". They were mutual organizations formed with the
stated purpose of producing a low risk debt instrument for
consumer's portfolios.19 The second were the Building
Societies, the forerunners of the Trust and Loan Companies.
Both of these groups sought to take advantage of certain
restrictions placed on the chartered banks. Namely, the
6% maximum interest on loans and the prohibition of
lending for mortgages. The legal status of these organ-

izations was established in an act of 1855 which clarified

 

l7Breckenridge, S stem, p. 58.
l8Breckenridge, History, pp. 36-37.

l9Canada, Royal Commission on Banking and Finance,
Report (Ottawa: Queen's Printer, 1965), p. 147.

 

118

their right to borrow via debentures and deposits and to
lend for mortgages.20 However, the status of their money
production at this time is unclear. Both deposits and
debentures paid interest and at least in several instances,
deposits were checkable up to a specified maximum.21

In 1850 the provinces of Upper and Lower Canada
were combined into the Province of Canada. The output
restrictions of the chartered banks, triple their paid—in
capital, had resulted in the opinion being expressed that
the supply of money was much too small to facilitate the
transactions of the Province.22 Rather then allowing the
existing chartered banks to increase their capital, and
thus their output maximum, a "free banking act" was
passed. The provisions of this act were: 1) the pro—
duction of notes less then five shillings was prohibited.
The production of notes of more than five shillings was
restricted to "banks". 2) the barriers of entry into the
money production industry were nominally removed.23

The Free Banking Act of 1850 extended the privi—
lege of note issue to "other persons or corporations" which

1. had a minimum of 25,000 6:0f capital stock,

2. had but one office, no branches.

 

20Ibid., p. 173. 21Ross, Bank of Commerce p. 220.

22Walker, A History of Banking in Canada, p. 42.
23Breckenridge, System, p. 106.

 

119

3. would deposit 25,0000? of provincial securities
with the Receiver General as a redemption fund
against possible default.zh
This act provided for free banking in name only.
Because of the severe restrictions, only five "free" banks
were formed and only one lasted beyond the Confederation.25
In 1866 the Free Banking Act was repealed and Canada re—
verted back to the old system of chartered banks.26
The result of this experience with "free banking"
was that the position of the chartered bank cartel was so
enhanced that it has not been challenged to this day. The
failure of the "free banks" was inevitable. Superficially
it seemed as though the cartel was willing to accept
competition in return for the legal restriction of note
issue. But the barriers to entry were never really lowered.
The restrictions placed on the "free banks" were tailor made
so that success was almost impossible. Thus, in 1866, the
chartered banks could point to themselves as the only pro-
ducers who could profitably produce notes, and they were
the only institution which had the legal right to do so.

The government first began to consider the pro—
duction of money in 1860. At this time, there was a

proposal to establish a national bank of issue, or

 

241b16., pp. 106—107. 25Ibid., pp. 113—114.
26161d., p. 116.

 

. uw—ww—‘v [7 a.

my

 

 

 

 

120

Treasury Department. It was defeated. However, in 1866,
the Province, being in financial trouble, decided that it
would use the profits from money production to finance its
own operations.27 Thus, the Provincial Note Act of 1866
authorized the entry of the government into the money
producing industry.

The exact beginnings of money-debt production is
unknown. One source cites the Agricultural Bank of Toronto
paying interest on deposits in 1834. This bank is described
as a joint stock company and possibly outside the cartel.28
Apparently, the cartel members were at first hostile toward
this practice29 but, by l8hl, the public statements of the
chartered banks contained the entry "cash deposits bearing
interest" without comment.30 In any event, the tradition,
and competitive necessity, of paying interest on certain
deposits was firmly established long before the confed-
eration.

In 1866, on the eve of the Confederation, the
structure of the modern Canadian money producing industry
was beginning to take shape. All of the modern producers

but two, the credit unions and the Provincial Savings

 

27Ibid., pp. 124-137.

28A. B. Jamison, Chartered Bankin in Canada
(Toronto: Ryerson Press, 1953}, p. 6.

29Ibid.

30Walker, History of Banking in Canada, p. #0.

 

121

Offices, were in operation. The underlying causes of

future difficulties in the determination of the amount of

money produced, the lack of a single regulatory body and

the tradition of money—debt production, were also present.

Briefly, the Canadian money producing industry as

it stood in 1866 is as follows:

I. 19 chartered banks.

A.
B.

Issued notes.

Accepted deposits.

1. Subject to checking.

2. A portion bore interest.

No explicit reserve requirement in dominant
money.

Output limit.

1. Total "liabilities" cannot exceed three
times paid—in capital.

2. Composition of "liabilities" unimportant.

II. Government production just beginning.

III. Savings Banks and Building Societies.

A.
B.

C.
D.

Could not issue notes.

Checkability of deposits unknown, but
probable.

Paid interest on deposits.

No output limit.

 

APPENDIX II

"ACTIVE BANK DEPOSITS

The Bank of Canada compiled and published, from
1946 to 1957, a series entitled "Active Notice Deposits."
Superficially, the rationale and the method for the con-
struction of this series bears a striking resemblance to
the approach used in this study. But, on further ex-
amination, the concept of money implied by this series is
very different from the one employed in Chapter II.

"Active Notice Deposits" was derived from total
"Notice Deposits", Personal Savings plus Other Notice
Deposits, and purported to take into account the low turn—
over of Notice Deposits.l The Bank of Canada felt that,
because the number of checks written on Notice Deposits was
far below that written on Demand Deposits, some correction
should be made so not to overstate the contribution of
Notice Deposits to the money supply. They recognized the
fact that, because of the explicit interest payments,
Notice Deposits were also held as a non-money asset. They
wished to include only the money portion in the money supply.

They reasoned that the money portion of Notice Deposits

 

1The information concerning the methods and motives
for this series was obtained from a number of interviews with
W. E. Scott, the Inspector General of Banks, and his staff
and the Research Department of the Banks of Canada during
December, 1967.

122

 

123

would have checks written against them, be "active," and
that the remainder would remain stationary, "inactive," in
the accounts.

The method used by the chartered banks for the
computation of interest payments on Notice Deposits allows
accurate estimation of "inactive Notice Deposits" to be
made. Actual interest payments are computed by applying the
announced rate and the actual interest payments, the mini-
mum balance, the amount against which no checks were written
in that period of time, can be computed. Subtracting the
minimum balance from total Notice Deposits, "Active Notice
Deposits" is obtained.2

Reformulating the Bank of Canada's analysis using
the terminology of Chapter II: abstracting from costs,
"Active Notice Deposits" earn only an explicit rate of re—
turn equal to the announced rate rd. Presumably, in
equilibrium, rm = rd 3 r, all rates equal the market rate.
But, at least in recent history, the announced rate has
been below the market rate. Therefore, in order for the
minimum balances, the "inactive" portion, to be held in
wealth portfolios, an additional stream of income must

accrue to them in order to bring their total rate of

 

2The fact that some Notice Deposits are non-check-
able tends to overestimate "Active Notice Deposits." Normal
cash withdrawals from a non-checkable account would result
in a dicrease of minimum balances in the same way as writing
a chec .

 

 

 

12h

return to approximately the level of the market rate. The
hypothesis will be set forth that this additional rate of
return is a money rate, equivalent to rm. Thus, a part of
minimum balances should be included in the money supply
even though they were not spent in the unit of time.

The concept of money implied by Bank of Canada's
rationale has some very strange characteristics. In es—
sence, the capital good money is defined in an ex post
sense. A dollar of Notice Deposits is included in the
money supply only if a check has been written against it,
if it has been used to facilitate exchange, in a period of
time. If this approach would be applied to other capital
goods its lack of generality would be obvious. The supply
of automobiles would include only those driven in a period
of time. A dollar bill would not be money if it remained
in a person's wallet or safe.

The problem arises because this definition of a
capital good depends on the unit of time chosen. If an ob—
ject is used for a certain purpose during this period of
time, it is a capital good. If it remains idle, it is not
a capital good, even if it were used in the previous period
and will be used again the following period. Most capital
goods are not in continuous use. By their nature they are
used only when needed, resulting in periods of idleness.
Simply because a good is not used in one period of time does

not imply that it is not capable of use, that it is not a

 

@133!" ‘-

 

 

125

capital good. Periods of activity and idleness vary be-
tween capital goods. Short of infinity, no one unit of
time can be chosen such that all capital goods are in use.
Rather, the definition of a capital good should not be in
terms of its use during one period of time.

Economics is often concerned with the change in
behavior associated with the possession, or non-possession,
of certain capital goods. Defining money in terms of one
of its effects in a period of time results in an incomplete
‘ view of the consequences. For example, an increase in the
quantity of money may increase the amount of money used in
transactions. But, part of the increase may be held for
later expenditures or as a permanent addition to a wealth
portfolio. By only looking at the amount of money that is
spent, one only sees that portion of the increase that has
been allocated to a particular use during this period of
time. Only one form of behavior, spending, is allowed to
be affected by a change in the quantity of money.

Despite the fact that "Active Notice Deposits"
imply a narrow concept of money, the originators of this
approach must be given credit for being the only ones to
attempt to adjust for the debt component of money—debt.
They intuitively saw that, because these deposits bore an
explicit return, they would be held in wealth portfolios
both as money and an interest bearing asset. The problem

arose because of their attempt to measure the money

 

 

 

126

component. Their definition results in measuring the portion
of the money component which has been used in a certain way.
It does not measure the money component itself. In other
words, "Active Notice Deposits" measure the result of be-

havior, not the magnitude which influences behavior.3

 

3There is no reason to believe that two identical
capital goods would be used with the same intensity. Thus,
the large differences between the turnover rates of Demand
Deposits and Notice Deposits does not, by itself, imply that
Notice Deposits are not entirely money. Even after adjusting
for the inactive portion, the turnover rates of "Active No-
tice Deposits," that portion which Egg used as money, is still
much below that of Demand Deposits.

TABLE 18
RATES OF TURNOVER

 

 

 

 

 

 

 

 

 

Active Notice De csits

Year Demand DepOSits Notice Deposits (Total?
1963 30. 39 7.55 1.20
1966 31.16 7.31 1.16
T965 31.69 7.20 1.19
1967 28.65 6.63 1.17
1968 29.86 7.27 1.16
1969 36.02 7.35 1.13
1950 33.00 8.29 1.27
1951 35.06 8.79 1.37
1952 35.19 8.16 1.32
1953 36.6 9.22 1.51
1956 38.11 8.65 1.39
1955 38. 53 9.02 1.44
1956 63. 01 9.66 1.66
Source:

Bank of Canada Statistical Summar , 1966, 71956.
Checques Cashed in Canada, D. B. 3.

Canada, Bank of Canada, Statistical Summarg (Ottawa: Bank
of Canada, 1966 and 195 65.

Canada, Dominion Bureau of Statistics, Che ues Cashed in
Canada, 1966 (Ottawa: Queen's Printer, 1957}, po 32.

 

 

 

APPENDIX III

AVERAGE OF MARGINAL RATE ?

Most Canadian money producers compute interest pay—
ments on money-debt by applying the announced rate to the
minimum balance in a period of time. This practice may be
though to create problems because the rate on moneydebt
used in Chapter II is computed by taking the interest pay-
ments and dividing by the average balance in a period of
time. Thus, the rate used is an average rate which, at
first glance, may not seem to be equal to the marginal rate.

Let us divide money-debt into two portions: (1) the
minimum balance in a period of time, and (2) the "active"
portion. Abstracting from explicit costs, the first earns
interest equal to the announced rate, the second earns no
explicit interest.

The "active" portion of money-debt earns no in-
terest and is obviously money. The only stream of income
is that obtained from its use as money.

Assume that the announced rate on money-debt is be-
low the market interest rate. Then, the minimum balances
earn a rate below its alternatives. The differences be-

tween the market rate and the announced rate must be

127

 

128

caused by the balances earning a money rate. As was dis-
cussed in Appendix II, money does not have to be spent in
order for it to be money. An amount of the minimum balance
equal to the discounted present value of the money stream
of income is just as much money as are the "active" bal-
ances. For the lack of a better term, the portion of mini-
mum balances which are money may be called "idle balances."
Or, they may be used to satisfy the "precautionary demand
for money."

It is not necessary to divide money-debt into
"active" and minimum balances in order to compute the
quantity of money. Assume $1000 of money—debt, an an—
nounded rate of 2%, a market rate of 5%, and the minimum
balance of $800. Three—fifths, $680, of the value of the
minimum balance must be money. When the "active" portion
is included, the total quantity of money is $680. The ef-
fective rate on money—debt (.O2(800)/1000) is 1.6%. From
this it would be concluded that (l — 1.6/5) = .68 or $680,
of money—debt is money. Both methods give the same answer.

This analysis implies that the difference in rates
applies to the changing of the form in which money is held.
The marginal cost of switching from "idle" to active bal-
ances, and thus lowering the minimum balances, is equal to

the announced rates.

 

SELECTED BIBLIOGRAPHY

Data Sources

Canada, Bank of Canada. Statistical Summary. Ottawa: Bank
of Canada, 1966 and later years.

. Research Department. Personal Interviews and
Correspondence. 1967-1968.

Canada, Department of Agriculture. Credit Unions in Canada.
Ottawa: Department of Agriculture, 1966 and later

years.

Canada, Dominion Bureau of Statistics. The Canada Yearbook.
Ottawa: King's Printer, 1938 and later years.

Canada, Inspector General of Banks. Personal Interviews and
Correspondence, December, 1967.

Canada, Superintendent of Insurance. Report of the Superin-
tendent of Insurance for Canada: Loan and Trust
Companie . Ottawa: Queen's Printer, 1965.

de Melto, Dennis. The Supply of Money in Canada 1867—1261.
Unpublished M. A. thesis, cGill University, 19 3.
La Federation Des Caisses Populaires Desjardins. Memoire.

LeVis: La Federation Des Caisses Populaires
Desjardins, 1967.

 

. Levis, Quebec. Personal Correspondence. January-
July, 1968.

Ontario, Registrar of Loan and Trust Companies, Report.
Toronto: Queen's Printer, 1938 and later years.

Quebec, Quebec Bureau of Statistics. Quebec Statistical
Annual. Quebec City: Quebec Bureau of Statistics,
1960 and later years.

Saskatchewan, Department of Co—Operation and Co-Operative
Development. Credit Unions. Regina: Lawrence Amon,
1958 and later years.
. Personal Correspondence. January-July, 1968.

129

"Tn.-;‘:’7A ‘-e " . -'

 

 

 

130
Theoretical and Institutional Background
B. C. Credit Union League. A Brief to the Royal Commission

on Banking and Finance. Vancouver: B. C. Credit
Union League, 1962.

Breckenridge, Roeliff Morton. The Canadian Bankin S stem 1817-
1890. Toronto: Printer Unknown, 1896.
. The History of Banking in Canada. Washington:
Government Printing Office, 1910.

Canada, Bank of Canada. Evidence of the Governor Before the
R0 a1 Commission on Bankin and Finance. Ottawa:
Bank of Canada, 1966.

Canadian Bankers Association. Submissions to the R0 al Com-
mission on Bankin and Finance. Toronto: Canadian
Bankers Association, 1963.

Canada, Royal Commission on Banking and Finance. Report.
Ottawa: Queen's Printer, 1966.

 

. Report, Apppndix Volump. OttawazQueen's Printer,

 

Credit Union League of Manitoba, and Co—Operative Credit
Society of Manitoba. Brief Presppted to the Royal

Commission on Bankin and Finance. Winnepeg: Credit
Union League, 1962.

Graham, J. F., Inman, M. K., Peitchinis, S., Reuber, G. L.,
and Taylor, J. C. The Role of the Trust and Loan
Companies in the Canadian Econogy. London: School
of Business, University of Western Ontario, 1965.

 

Hay, Keith. "Money and Cycles in Post Confederation Canada."
Journal of Political Economy. 75 (June, 1967),
260-275.

Jamison, A. B. Chartered Banking in Canada. Toronto: Ryerson
Press, 1953.

Johnson, Joseph French. The Canadian Banking System. wash-
ington: Government rinting Office, 91 .

La Federation Des Caisses Populaires Desjardins. Brief Sub-
mitted by La Federation de Quebec des Unions Re ion-
ales des Caisses Populaires Desjardins to the Royal

Commission on Bankin and Finance. Levis: La Feder-
ation Des Caisses Populaires Desjardins, 1962.

 

131

Patterson, E. L. Stewart. Canadian Bankin . Toronto: The
Ryerson Press, 196 .

Pesek, Boris P., and Saving, Thomas R. Mone Wealth and
Economic Theory. New York: The Macmillan Company,l967.

- .The Foundations of None and Bankin . New York: The
Macmillan Company, 1968.

Ross, Victor. The History of the Canadian Bank of Commerce.
Toronto: Oxford University Press, 1920.

Trust Companies Association of Canada. Submission to the Royal

Commission on Banking and Financ . Toronto: Trust
Companies Association of Canada, 1962.

 

Walker, B. E. A History of Banking in Canada. Toronto: Printer
Unknown, 1920.

The Demand for Money

Chow, Gregory. "On the Long Run and Short Run Demand for Money."
Journal of Political Economy. LXXIV (April, 1966),
Ill—132.

Feige, Edgar. "Expectations and Adjustments in the Monetary
Sector." American Ecopgmic Review. LVII (May, 1967).
662-673.

 

Friedman, Milton, and Schwartz, Anna. A Monetary History of the
United Statgp. Princeton: Princeton University Press,

19 30

Hamburger, Michael J. "The Demand for Money by Households,
Money Substitutes, and Monetary Policy." Journal of
Political Economy. LXXIV (December, 1966), 600—626.

Laidler, David. "Some Evidence on the Demand for Money."
Journal of Political Economy. LXXIV (February, 1966),
55‘690

. "Rate of Interest and the Demand for Money."
Journal of Political Economy. LXIV (December, 1966),
563-551.

Laumas, G. S., and Formuzis, P. "The Demand for Money in
Canada." Canadian Journal of Economics. I (August,

1968), 6881699.

 

 

.-..ij 1.7..1- _

 

132

Macesish, George. "Determinants of Monetary Velocity in
Canada, 1926m1958." Canadian Journal of Economics
and Political Science. 28 (May, 19627j'265n256.

. "The Rate of Change in Money Stock as a Leading
Canadian Indicator." Canadian Journal of Economics and
Political Science. 28 (August, 1962), 626-632.

Mercure, Giles. Credit Unions and Caisses Populairep. Working
Paper for the Royal Commission on Banking and Finance.
Ottawa: Royal Commission on Banking and Finance, 1962.

Meltzer, Allen H. "The Demand for Money: The Evidence from
Time Series." Journal of Political Economy. LXXI
(June, 1963), 217~229.

Morrison, George R. Liquiditnyreference of Commercial Banks.
Chicago: University of Chicago PPess, 1966.

Musgrave, R. A._Tpe Theogy of Public Finance. New York: McGrawu
Hill Book Company, 1959.

 

Shapiro,9 Harold. "Distributed Lags, Interest Rate Expectations,
and the Impact of Monetary Policy: An Econometric
Analysis of a Canadian Experience." American Economic
ngggp. LVII (May, 1967), 663~680.

 

Teigen, R. "Demand and Supply Functions for M ney in the
United States." Econometrics. XXXII TOctober, 1966),

478-693.

Walters, A. A. "The Demand for Moneyw The Dynamic PrOperties
of the Multiplier." Journal of Political Economy.
75 ...ne, 1967), 292-300.

 

Zaremka, Paul. "Functional Form of the Demand for Money."
Journal of the American Statistical Association. 63

u“

(Jufie, 1968), 502-511.