THE GENERAL THEORY O F EMPLOYMENT,

INTEREST A N D MONEY’
It has become traditional for reviewers of important books
to begin by saying that it is impossible within the limits of a
single article to do more than consider very briefly a few of the
questions at issue. Such a statement would assuredly be justified
in the present case, for in this book Mr. Keynes attacks one of the
fundamental assumptions which has underlain orthodox theory
since the days of Ricardo. This is the doctrine which used t o be
expressed categorically in the phrase L ‘ Supply creates its own
demand.” Later writers have been more guarded on the subject,
and often refrained from stating it specifically in any form at
all. But however it might be expressed or implied, orthodox
theory has continued to be based on the principle that “ what
constitutes the means of payment for commodities is simply
commodities”z (Mill) ; from which it follows ( i n f e r ulia) that
money makes no difference except frictionally, that consumption
is limited by production and not vice versa that general over-
supply is impossible, and that, to quote Professor Pigou,
unemployment is due to the fact that “ frictional resistances
prevent the appropriate wage adjustments from being made
instantaneously.” In place of this ah. Keynes seeks to sub-
stitute a monetary theory of production according to which
unemployment may be due, not to labour’s refusal to accept a
lower reward, but to a deficiency of “effective demand.”
Probably the first thing which many people wanted to know
on the announcement of the new book was its relation t o the
Treatise on N o n e y . Xr. Keynes deals with this question in the
preface, and there is little need for me to add anything. There
have been important changes in method and terminology, the
latter being bound, I fear, to create much initial confusion.
But the underlying principles are fundamentally the same,
though they have been clarified and, in the process, considerably
developed. Those who have accepted the main thesis will, as
soon as they are accustomed t o the new terminology, find a more
penetrating an d more logical exposition of i t ; those who still
1. bhcmillan & Co. Ltd. 5/- atp. Pp. 403.
2. La- writera have, of eourne. added “rervim” t o eommoditier,: but that is
not the point a t issue.
2s
JUNE 1936 EMPLOYMENT, INTEREST AND MONEY 29

remain unconvinced will be stimulated to greater endeavours t o

refute the heresies, now more dangerous than ever as a result
of more forceful expression.
The most fundamental change is a clear r.ecognition that
the thing to be studied is the forces which determine the total
volume of output and employment, rather than the various
price-levels. Not a few people, including the present reviewer
and probably Mr. Keynes himself, must have felt that this
really was the quaesiium even in the Treatise, and that the
analysis was for that reason rather artificial. In effect the
process was supposed to work in two stages; an increase in
investment, for example, would lead first to higher prices being
realized for an unchanged output of consumption goods, and the
resulting windfall profits then provide an inducement to entre-
preneurs t o expand output. Mr. Keynes was assailing the notion
that money was a mere counter which could exert at the most a
transitory influence on “ real ” things like output and employ-
ment; but the old tradition was so strong that he made i t work
through the medium of changes in prices, which would only
subsequently lead t o changes in output. To quote from the
preface of the new book: “ My so-called fundamental equations
were an instantaneous picture taken on the assumption of a given
output. They attempted t o show horn, assuming the giv,en out-
put, forces could develop which involved a profit-disequilibrium,
and thus required a change in the level of output. But the
dynamic development, as distinct from the instantaneous picture,
was left incomplete and estremely confused. )’
Of all the ideas contained in the Treatise the most familiar
is that associated with “ the difference between saving and
investment.’’ This phrase has become a catchword which has
been applied, or rather mis-applied, to all manner of problems
without any reco-pition of the special meanings attached t o the
terms. The concept depended essentially on the definition of the
entrepreneurs’ income (and hence of their saving) ; this mas
taken as being neither the profit actually realized from current
operations, nor the profit which they expected when they decided
on what scale to conduct these current operations, but in some
sense as a “ normal ” ur “ standard ” profit. As a result
saving exceeded investment by the excess of this normal profit
over that actually realized; in effect the entrepreneurs were
considered to draw the difference from a sort of contingency
fund before computing their income, and were correspondingly
bound to add any surplus t o it. Their saving was then the
30 THE ECONOMIC RECORD JUNE

difference, positive or negative, between the income so computed

and their expenditure on consumption.
Now this concept has certain advantages mhere the entre-
preneur is a joint-stock company, which will probably put any
abnormal profits to reserve before declaring a dividend, and may
draw on reserves to maintain a conventional rate. B u t it is a
somewhat artificial notion, and in the new book it has been aban-
doned, income being measured by the profit actually realized.
There are certain complications on minor matters, notably the
question of depreciation; but the important point is that total
income of the community for any period is simply the value of
the output of that period. Saving is defined as before as the
difference between income and expenditure on consumption ;
investment is the part of output which is not consumed. It
follows as a matter of simple subtraction that saving and invest-
ment, as now defined, are necessarily always equal; this is
inevitably true whether the supply of money is being altered or
not, whether we are a t the bottom of a slump or at the height of
a boom.
This contains, of course, nothing fundamentally new; it
is simply the espression in the new terms of the proposition in
the Treatise I = S + - Q the value of investment is equal to
saving plus windfall profits. B u t a t first sight it may appear
difficult to reconcile with, for example, the freedom enjoyed
by each individual to save as much of his income as he likes.
The solution is, however, quite straightforward. When we think
of an individual deciding how much to s i r e we instinctively take
his income as given; his actions are too insignificant to have
any reactions on it. But for the whole community this is, in
general, untrue. If, for the moment, we assume that the amount
of investment to be done during any period is fixed, then the
national income must so adjust itself that the community will,
in the aggregate, do just that amount of saving. This may come
about in various ways, depending on the type of economic
system in which we live. To take a simple example, imagine
a closed system i n which there is only one entrepreneur, and he
decides irrevocably what is to be produced and sells the whole
output of consumption goods to his employees for whatever it
will fetch. In this case the adjustment will be secured wholly
through changes in the eptrepreneur’s income. If the public
spend the whole of their wages, then he w i k m a k e a profit on
the sale of consumption goods equal to the eip+diture on invest-
ment; if the public save a part, then hi# takings, and so his
income and his saving, are correspondingly reduced.
 1936 EMPLOYMENT, INTEREST AND MONEY 31
In this example the adjustment is effected solely by changes
in prices altering the income of the entrepreneur without affect-
ing his expenditure on consumption. But as a rule there wiIl also
be changes in income due to changes in employment, and these are
frequently responsible for the bulk of the adjustment. An
increased desire to save will not merely lead to entrepreneurs
wceiving a reduced income, but also in a diminished volume of
employment, and the extra saving of those still employed will be
balanced in part by the reduced (probably negative) saving of
those thrown out of work. Another consequence will probably
be an accumulation of stocks,.which represents unconsumed out-
put, and, therefore, constitutes investment. The reader must on
no account think of imestment as applying solely t o construction
of k e d capital; the necessary equality between saving and
investment may not infrequently be secured partly through
changes in the size of stocks.
P u t broadly, Mr. Keynes’s new analysis may be summed up
as follows. We visualize a series of levels of employment. As
we go up the scale the community’s real income increases. Its
psychology is such that consumption will increase with increased
income, but by a smaller amount ; the exact relationship is given
by a function which Mr. Keynes calls “ the propensity to con-
sume. ” Hence employers would make a loss if the whole of the
increase in employment were devoted t o production for immedi-
ate consumption; the increase will only be justified if part of it
is engaged on producing investment goods. Given the propensity
to consume the level of employment is determined by the amount
of current investment; for it must give that level of income out
of which the community will do this amount of saving.
It remains, therefore, to find out what determines the amount
of current investment. This, according to Mr. Kepes, depends
on the relation between ‘ ‘ the schedule of the marginal efficiency
of capital ” and the rate of interest (or more strictly, the com-
plex of rates applicable to debts of different maturities). The
first is constructed by considering the supply price and prospec-
tive yield over its whole life for each possible type of new capital
asset. The marginal efficiency of that type of capital is then
defined as being equal to “ that rate of discount which would
make the present value of the series of annuities given by the
returns expected from the capital asset during its life just equal
to its supply price ” (p. 135). Now if there is increased invest-
ment in that type of capital during any period the prospective
yield will fall, and the supply price may rise owing to pressure
32 THE ECONOMIC RECORD JUNE

on the facilities for producing it. So that we can build up a

schedule showing how much new investment there will have
t o be in it to reduce its marginal efficiency t o any given figure.
This may be done for all types of capital; the results are then
aggregated to give a schedule showing the total amount of invest-
ment there will have t o be to reduce the marginal efficiency of
capital in general to any figure. Mr. Keynes quite rightly
emphasizes that it is the ezpected yield and the current supply
price of newly-produced assets which are relevant; since it is
obvious that these expectations cannot be quickly verified o r
disproved, it follows that the state of confidence is vitally
important.
Assuming for the moment that the rate of interest is given,
Mr. Keynes deduces from this schedule what the amount of
investment will be; for investment will be pushed to the point
which reduces the marginal efficiency of each sort of capital to
approximate equality with it. Without disputing the general
validity of the method, I feel that this part of the book needs
more extended treatment, particularly in the matter of risk and
the varying estimates made (often for good reasons) by different
people. Mr. Keynes inserts a very helpful discussion of the two
sorts of risk which are relevant. There is firstly entrepreneur’s
risk, or the doubt in his mind as to whether he will actually earn
the yield he considers most probable ; where he is using his own
money, this is the only relevant one, and presumably (though
BIr. Eeynes does not explicitly say so) me deduct an allowance to
cover it from the expected yield in calculating the marginal
efficiency of capital. But there is also another risk where the
entrepreneur uses borrowed capital ; the entrepreneur will not
invest unless the probable yield esceeds the interest he has to
pay by enough to cover his risk; but the lender mill not lend
unless the rate he receives exceeds the “ pure ” rate by enough
to cover the risk of default, voluntary or involuntary. The
allowance for this “ lender’s risk ” is a sort of handicap against
the entrepreneur who has t o borrow; and in addition there is
the cost of bringing borrower and lender together, which may be
considerable.
Xow in constructing his schedule Mr. Keynes seems t o take
no account of differences between individuals. He talks of the
“prospective yield’’ as if it had only one possible value, gives
no specific instructions as to the treatment of risk, and compares
the resulting marginal efficiency with the pure rate of interest
1936 EMPLOYMENT, INTEREST AND MONEY 33
for loans of the relevant maturity.a To make the treatment
logical we must do thc thing in considerably greater detail. If
the asset under consideration is not readily marketable, then we
consider each potential owner individually. We deduct from the
yield, as estimated by him, firstly entrepreneur’s risk, and
secondly lender’s risk and the cost of securing the loan (where
applicable) ; and then we relate this t o the supply price.
(Alternatively we can add n capital sum t o the supply price t o
cover these; this is perhaps better for the I ‘ preliminary
expenses” involved in securing the loan.) By combining the
results for different individuals, not forgetting that one may
inv,est in several units, we can build up a schedule which really
can be compared with the rate of interest. Where the asset
is one with a free market there are two additional considerations.
The individual who takes an unusually optimistic view can buy
an old unit rather than construct a new one ; and the initiative
is frequently taken by promoters, who are influenced more by
what. they think the public will think of the enterprise than in
their own views. To state precisely how the schedule of the
marginal efEciency of capital is to be constructed is almost impos-
s i b l e t h e r e are many complications and mutual inter-actions
which I have not space to describe.* But we need not reject the
general idea of a scale showing how much extra investment would
be considered payable if the rate of interest were lowered; it
appears in many theories under such titles as “ the demand
curve for capital.”
We are left then to find what determines the rate of interest.
It is here that Mr. Keynes differs profoundly from the exponents
of the classical theor?. Space is too limited for me to reproduce
or criticize his attack on this doctrine; it would indbed be a
waste of time, for I consider it well-founded, and the next move
should obviously come from the defence. The substance of it
is that in constructing their supply curve of savings the classical
economists tacitly assume that the level of employment is
unaffected by decisions as to saving-in effect that their theory
is only applicable to a world possessed of some mechanism
whereby involuntary unemployment is made impossible. Mr.
3. Thim in the procedure followed in the past deallng specifically with the
marginal efficiency of c:?ltal (cf. footnote on D. 137). But in a preliminary outline
he uaes the expression complex o f rat- of interest on h M of varioua maturitics
and rirks” (p. 28) (w italia).
4. Thus the schedule k gr u tly af?ected hp the level of employment. b e a u x
this r-ta on confidence. Investment rhould perhaps be rwreentad M a function
of income ( Y ) 8nd the mte of interat ( r ) n t h e r than of r alone. This anpean
to involve circular reaaoning. because we atarted by raping that income depend.
on invcatment. But, M ahown later, thin im not a valid objaction, the facton mutu-
ally determining one another.
C
34 THE ECONOMIC RECORD JUNE

Keynes’s “general” theory is designed to apply whatever the

level of employment, including the classical theory as a special
case.
This theory is built round the conception of “liquidity
preference ” as explaining man’s desire to hold cash. There is
firstly the familiar consideration that people need a certain
amount of liquid resources to assist them in spending their
income and carrying on their business. This demand for money,
so to speak, “on current account” arises essentially out of its
use as a medium of exchange and depends in the short period
on the Level of activity. But money is also to the individual a
store of value-a fact which is mentioned at the beginning of all
text-books on the subject, but seldom fully explored. Moreover,
it had several great advantages as a means of holding wealth.
It involves no carrying costs-indeed a modest return can be
secured by holding it in the form of a fixed deposit; it is liquid,
and involves no expenses of acquisition and realization; and in
contrast with bonds, its value cannot depreciate. An owner of
wealth has t o balance these advantages against the increased
income which he can get by holding a bond in decidbg what
proportion of it he will hold in the form of cash. The amount
will in general increase as the price of bonds rises (ie., as the
rate of interest falls), largely because of the added risk of capital
loss involved in holding bonds. Thus Mr. Keynes constructs a
second liquidity function which relates the amount of cash which
people wiLl want to hold, so to speak, “ on capital account ”
with the rate of interest. The total amount of money available
( M )is determised by monetary policy (or, in the absence of
policy, by the supply of gold or some other accident), and this
must be held by somebody, either on current account or on capita1
account. Hence we have the relation
Y = ‘I21 -+ ;I12 = L1 (Y)+ L2 (f)

Now if we know P, the level of income, we can deduce UL

and hence M 2; and we can then say that the rate of interest must
be such that people will want to hold this amount of money on
capital account rather than transfer to bonds. But the object
of finding the rate of interest is to deduce the amount of invest-
ment and hence the Ievel of income. Are we then reasoning in a
circle? The answer is, no, we are merely faced with the
inevitable difficulty of trying to describe a system where the four
variables mutually determine one another. This can best be seen
by a sort of mathematical shorthand, where I stands f o r Invest-
1936 EMPLOYMENT, INTEREST AND MONEY 35

ment, S for Saving, Y for Income and r for the rate of interest.
Our four propositions are represented approximately by
(1) s = f ( Y ) .
(2) I = g ( r ) . ( 8 )
(3) I = 8.
(4) M = L1 ( Y ) + L2 IT).

These are set down here to show that we have really got enough
relationships. Mr. Keynes, quite rightly in my opinion, depre
cates the spurious air of exactness introduced by too much mathe-
matics. But in his endeavour t o describe the system without
this sort of shorthand he has tended to obscure the fact that the
determination is mutual. This is particularly noticeable in the
first paragraph of p. 248, where he glosses over the fact that
the rate of interest depends on M 2 and not on Y when dealing
with investment.
And now I must make the reviewer’s inevitable apology f o r
having-attempted the almost impossible and failed. There are
any number of important things in the book which I have not
had space to mention. There is a useful chapter on the vexed
question of a reduction of money-wages. This would be more
convincing if it took account of the fact that many non-wage-
earners relate their spending to the money derived in t4e form
of dividends, etc., from the production of a previous period.
If this money is made more valuable there is a real reason why
their spending should be a greater proportion of their income
from current operations. And Mr. Keynes concludes with a
chapter on the social implications of the theory; but this is
rather like showing Moses the Promised Land which he can
never enter. A blissful picture is drawn o i a society which is
supplied with all the capital it needs and has reduced the rate of
interest t o zero; but a large part of the book has been devoted
t o showing the difficulty of doing this.
Parts of the book are undoubtedly difficult, but they are
relieved by less technical interludes, such as the magnificent
description of the operations of Wall Street. Moreover, Mr.
Keynes has a clarity of style which saves much confusion. If
the propounders of rival theories would only set out what they
take as given and what they seek to find, as he does in Chapter
18, then we should be spared many a headache. Many mill con-
sider that he under-rates the dangers of expansionism, but that
is surely better than the attitude of people who forbid the State
6. If we accept the argument of ths previous footnote. this rhould be written :
I = h (Y1 r ) .
36 THE ECONOMIC RECORD JUNE 1936

to increase employment for fear of ‘‘ distorting the structure of

production,” or because of the absence of any corresponding
L i
genuine saving.” The logic of the argument would be
improved if “ the rate of interest ” mere not used so frequently
to represent “ the cost of raising capital ”; particularly in Aus-
tralia the other elements, such as quantitative control of credit,
are often far more important, and the rates applicable to different
industries and borrowers may move differently for institutional
reasons. B u t whether or not one accepts the new doctrine, fen.
mill dispute Mr. Keynes’s remarks about the importance of the
questions at issue. To take a single example: “ A decreased
readiness to spend will be looked on in quite a different light if,
instead of being regarded as a factor which wdl. ceteris paribas,
increase investment, it is Seen as a factor which will, ceteris pari-
bus. diminish e m p l o p e n t ” (p. 185).
W. B. REDDXWAY.
Z‘niversity of Xelboiirne.