MICHAL KALECKI

The Marxian equations of reproduction

and modern economics *

Before we start dealing with the proper subject of this paper we shall

modify somewhat the Marxian division of economy into departments in

order to simplify our argument and in order to focus on the basic problem
of the reproduction schemes.
First, instead of including producer goods in Department 1, we will ,

assume that it covers the total value of gross investment inclusive of

the respective raw materials. Thus this department represents the inte-
grated production of all final non-consumer products. (We disregard
in our argument as does Marx - when he deals with reproduction sche-
mes - both foreign trade and government revenue and expenditure.)

Second, we treat likewise the consumer goods, i.e., we include in the

department which covers their output the production of respective raw
materials from top to bottom. Moreover, fully in the Marxian spirit,
we distinguish the following two departments : Department 2 producing
consumer goods for capitalists and Department 3 producing wage goods.
‘
&dquo;
We obtain thus the following &dquo; tableau 6conomique of the national
income where Pi, P2, P3 are gross profits (before deduction of depreciation)
in the respective departments, WI, W2, W3 - the respective wages; P and
W aggregate profits and wages, and finally I gross investment, C k - capi-
-

* This article was

presented as a background paper for the Symposium on the
influence of Karl Marx on contemporary scientific thought, Paris, May 8-10, 1968,
organized under the auspices of Unesco by the International Social Science Council
and the International Council for Philosophy and Humanistic Studies.

Downloaded from ssi.sagepub.com at FUDAN UNIV LIB on May 16, 2015

74

talists’ consumption, CW - workers’ consumption and Y - gross national

income (before deduction of depreciation).

. &dquo;
.. I’ !&dquo;
< : ;
, _
_

_
’:; It:>
n .. ’

It will be assumed, as Marx does, that the workers do not save. More-
over, we shall disregard the problem of possible piling up of stocks of
unsold goods as only a passing phenomenon. It is then easy to arrive
at the fundamental Marxian &dquo; equation of exchange &dquo; between Depart-
ments 1 and 2 on the one hand and Department 3 on the other.
Profits in the latter, P3, are materialised in the wage goods which are
left to the capitalists of that department after payment of wages W3 which
absorb an equal amount of wage goods. Thus the wage-goods of the
value P3 are sold to the workers of Departments 1 and 2, that is :

Marx considers this equation in the context of expanded reproduction

proceeding at a given constant rate r. It is easy to see, however, that the
equation holds good under all circumstances as long as there is no piling
up of stocks of unsold goods, as mentioned above.
Considered in this general context equation (1) leads to a proposition
that -

given the distribution of income between profits and wages in the

three departments investment I and consumption of capitalists Ck deter-
-

mine profits and the national income. Indeed, let us add PI + P2 to both
sides of equation ( 1 ). We obtain :

w ’&dquo;’&dquo; &dquo; ..

Hence : ’’&dquo;~- ’

Wl W2 W3 b .

I obtain
b .
Moreover, if
.f we d
denote
wl,
i k Cw
by wl,
W_z, w_3 W2, w3 respectively, we

from equation ( 1 ) : i..... - .. -

have for the of wage ~~‘v’ ’&dquo;

Consequently, we consumption goods : -

Downloaded from ssi.sagepub.com at FUDAN UNIV LIB on May 16, 2015

 75

and for the national income : ,

Thus the national income (or product) Y which can be sold and the
profits P which can be realised are determined in all circumstances (and
not only in a state of uniformly expanding reproduction) by the level of
investment I and consumption of capitalists Ck (given the distribution
of income between wages and profits). A question may be raised as to
why equations (2) and (4) must be interpreted in this way and not the
other way around, i.e., that investment and consumption of capitalists
are determined by profits and national income. The answer to this rather
crucial query is as follows.
Investment and capitalists’ consumption in the short period consi-
dered are the outcome of decisions taken in the past and thus should be
considered as given. With regard to investment, this follows directly
from the time-lag dependent on the period of construction. But changes
in capitalists’ consumption also follow those in profits with some delay.
Now, sales and profits in a given period cannot be a direct outcome of past
decisions : the capitalists can decide how much they will invest and consume
next year but they cannot decide how much they shall sell and profit. The
independent variables in a given period are investment and capitalists’
consumption. It is these magnitudes that through the equations (2) and
(4) determine the levels of national income and profits which can be rea-
lised.

’
,’
- .. , ,’ ;
_ _

-.’ -~’
m

The decisions of capitalists with regard to their investment and consump-

&dquo;
tion are made in real &dquo; rather than in money terms, that is I and Ck should
be calculated in stable prices. If wl, w2, w3 are constant and money wage
rates in all three departments change in the same proportion, the same
is true in this case of prices of the produce of these departments. More-
&dquo;

over, as is easy to see, equations (2) and (3) will hold also in real &dquo; terms.
&dquo;

Any increase in real &dquo; investment or capitalists’ consumption results

under these circumstances, in an increase in output of Department 3, Cw
to provide for a surplus of this department P3 sufficient to meet the demand
generated by the higher wage bills in Departments I and 2, i.e., W2 + W3.
However, such repercussions of an increase in I or Ck are obviously
possible only if there exist unused capacities in Department 3. Imagine
that such is not the case. Then C, is fixed in real terms, i.e., is equal to a
constant B. In this case the increase in money value of Wi + W~ will
cause a rise in prices rather than in production of wage goods. The result
will be that the &dquo; real &dquo; value of W1, W~ and W3 will be reduced as compared
with the levels which would be achieved if unused capacities existed in

Downloaded from ssi.sagepub.com at FUDAN UNIV LIB on May 16, 2015

76

w W2 w w
Department 3. Consequently w,
= II ’ W2
=

,W3
C, .~C, = -~,
B
where
=

&dquo;
all magnitudes involved are to be now interpreted in real &dquo; terms, will
decline in the proportion reciprocal to the increase in the prices of wage
goods. Equation (3) can now be written in the form : -

When I and/or Ck decrease, wl, w, and W3 decline in such a proportion

as to render the left hand side of the equation equal to B 1.
Sections II and III represent in fact the gist of the modern theory of
effective demand. As will be seen, this theory may be derived in full
from the Marxian equation (1) representing the exchange between Depart-
ments 1 and 2 on the one hand and Department 3 on the other, if this
equation is considered in the general context rather than in that of uni-
formly expanding reproduction.

IV

Let us now turn to the significance of the equations (2) and (4) just
in the latter context, i.e., in the process of a uniform accumulation of capi-
tal. Let us denote the &dquo; real &dquo; stock of capital by K, the rate of net accu-
mulation by r and the rate of depreciation by 8. In this case we may write
the &dquo; equation of accumulation &dquo;, recalling that I is investment gross of
depreciation, in the form :

Since we are considering the long-run process of growth, let us postulate

that capitalists’ consumption Ck is proportional to profits P. Since
according to formula (2) the latter are equal to I + Ck it follows that Ck
bears a constant relationship to I. We thus have :

In consequence we may write equation (4) in the form :

and substituting in it for I its value from equation (5) we obtain :

1. In a socialist economy prices of consumer goods are always fixed relative to wages
in such a way as to secure a full utilisation of the productive capacity B, i.e., the equation

1 Wl - W9 I
I - w
= B is permanently fulfilled (Ck obviously equals 0 in this case).

Downloaded from ssi.sagepub.com at FUDAN UNIV LIB on May 16, 2015

 77

The national income Y thus bears a constant relationship to the stock

of capital K (provided that WI’ w,, W3 do not change 2). With a given
relationship of productive capacity to the stock of capital the degree of
utilisation of equipment is constant. Thus if capital equipment is
satisfactorily utilised in the initial position, this state of affairs is main-
tained in the course of expanded reproduction and the problem of
effective demand does not arise.
It is this approach that is inherent in many contemporary theories
of economic growth. In particular if we differentiate the equation (7)
we obtain : ]

Now, with
so-called
a constant satisfactory utilisation of equipment,
ratio which we denote by R. Moreover, r K is
dY is the

capital-output
r K
the net investment and thus
the national income which
rY
is the relative share of accumulation in
shall denote We thus have :
we by a.

or

which is the basic formula of the Harrod-Domar theory (in which, however,
the coefficient a represents the &dquo; propensity to save of the population &dquo;
rather than the ratio of net accumulation out of profits to the national income
which depends on its distribution between capitalists and workers).
In fact many of the contemporary theories of growth are simply
variations on the theme of Marxian schemes of expanded reproduction
which are represented in this paper by equations :

and

v r

The repercussions of changes in investment and capitalists’ consump-

tion described in section II do not raise, I believe, any major misgivings.
In contrast to this, the moving equilibrium described in section III depends
on the very far-reaching assumption that capitalists are willing to engage

2. If the productive capacities of all three departments expand at the same rater
the shortage of wage-goods discussed in the preceding section will not come into the
picture.

Downloaded from ssi.sagepub.com at FUDAN UNIV LIB on May 16, 2015

78

in investment which increases their capital at a constant rate r per annum.

What happens, however, if having become more cautious (perhaps under
the influence of a change in the social structure of their class) they decide
to reduce investment from (r + 8) K to (r’ + 8) K where r’ < r?
y

.
It follows directly from formula (7)
that §
and thus the degree of utilisa-
. r’ + 3
tion off.
equipment declines
d I. ’h
in the I off the
a result h decline
d I’
proportionr+r +0 as

of effective demand. It is clear that in this situation the &dquo; cautious &dquo; capi-
talists will not be any more agreeable to a lower rate of accumulation r’
but will reduce it further to r&dquo; < r’, and this will in turn affect correspond-
ingly the degree of utilisation of equipment.
Some economists tend to consider this phenomenon as a down-swing
phase of the business cycle which takes place around the initial path of
growth. However, such a proposition is not well founded : there is no
reason why having left the initial unstable path, investment must fluctuate
around it rather than around the depreciation level 8K. Or to put it in
Marxian terms : why cannot a capitalist system, once it has deviated down-
wards from the path of expanded reproduction, find itself in a position of
a long-run simple reproduction ?
In fact we are absolutely in the dark concerning what will actually
happen in such a situation as long as we have not solved the problem of
determinants of investment decisions. Marx did not develop such a theory
but nor has this been accomplished in modern economics. Some attempts
have been made in the development of the theory of cyclical fluctuations.
However, the problems of the determination of investment decisions invol-
ving the elements associated with the long-run trend are much more difh- &dquo;
cult than in the case of the &dquo; pure business cycle (i.e., in a system which
in the long run is subject to simple reproduction). I myself tried to do
something along these lines but I consider my work in this field to be defi-
nitely of a pioneer nature 3. One thing, however, is clear to me : the long-
run growth of the national income involving satisfactory utilisation of

equipment in a capitalist economy is far from obvious.

’
’

! ~
:
= . ..- , .
i

VI .
’
&dquo;’
_ , ,
v
’
-,

That Marx was deeply conscious of the impact of effective demand

upon the dynamics of the capitalist system follows clearly from this pass-
age of the third volume of the Capital : &dquo; The conditions of direct exploi-
tation and those of the realisation of surplus-value are not identical. They
are separated not only by time and space but logically as well. The former

3. A new paper of mine on the subject appeared in the June issue of the Economic
journal.

Downloaded from ssi.sagepub.com at FUDAN UNIV LIB on May 16, 2015

 79

are limited merely by the productive capacity of the society, the latter by
the proportions of various branches of production and by consumer power
of the society &dquo;.
However, he did not systematically scrutinise the process described
by his reproduction schemes from the point of view of the contradictions
inherent in capitalism as a result of the problem of effective demand.
It is one of his most prominent followers, Rosa Luxemburg, who express-
ed very definite and even extreme views on the subject : she rejected alto-
&dquo;
gether the possibility of long-run expanded reproduction if no external
markets &dquo; are in existence. By &dquo; external markets &dquo; she understood
those outside the world capitalist system consisting not only of underdevel-
oped countries but also of the non-capitalist sectors of developed capitalist
economies, for instance, peasant agriculture as well as government pur-
chases.
Her argument suffers from the fact that she considers investment
decisions as being made by the capitalist class as a whole and this class is
frustrated by the knowledge that finally there is no market for the economic
surplus. However, her scepticism as to the possibility of long-run expanded
reproduction is valuable because the self-propelled growth of capitalist
economy cannot, indeed, be taken for granted. If this economy expands
&dquo;
at all without the assistance of external markets &dquo;, this, to my mind, is
due to certain aspects of technical progress which, however, do not neces-
sarily assure a satisfactory long-run utilisation of equipment.
&dquo;
Nor should the significance of external markets &dquo; in the develop-
ment of capitalism be disregarded. In particular, in present-day capita-
lism the &dquo; external markets &dquo; in the form of government expenditure,
especially on armaments, play an important role in the functioning of capi-
talist economies. This expenditure to the extent that it is financed by
loans, or even by taxation of capitalists, contributes to the solution of the
problem of effective demand because its effect is not offset by the decline
in investment and consumption. (The latter would be the case if this
expenditure were financed by indirect or direct taxation of workers.) Thus
today the &dquo; external markets &dquo; in this particular form are even of greater
significance for expanded reproduction than at the time when Rosa Luxem-
burg propounded her theory.
The high degree of utilisation of resources resulting in fact from these
&dquo; &dquo;

government-made external markets has a paradoxical impact upon Wes-

tern economic theory. It creates an atmosphere favourable to the cons-
&dquo; &dquo;
truction of models for the growth of laissez faire capitalist economies
unperturbed by the long-run problem of effective demand.

Michal Kalecki is Professor of Economics at the Central School of Planning and Sta-
tistics in Warsaw, and a member of the Polish Academy of Science. Among his major
: Essay on the theory of the business cycle )
works are 1933 (
( ); Theory of
in Polish
economic dynamics ); 1954 Outline of the theory of growth in a socialist economy
(
"

1963 (
(
) in Polish
); Trend and business cycles reconsidered ", Economic journal,
June 1968.

Downloaded from ssi.sagepub.com at FUDAN UNIV LIB on May 16, 2015