Capital accumulation and the composition of capital

Abstract

This paper reviews the Marxian concept of capital accumulation in the light of Paul Zarembka's (2000) recent contribution, taking into consideration the concepts of competition and composition of capital. It shows that accumulation is best understood through a class analysis. However, the capital relation is influential at different levels and it encompasses a broad range of features of modern society. In this context, this paper proposes a richer and more encompassing analysis of accumulation.

Full text

CAPITAL ACCUMULATION AND THE
COMPOSITION OF CAPITAL
Alfredo Saad-Filho
ABSTRACT
This paper reviews the Marxian concept of capital accumulation in the light
of Paul Zarembka’s (2000) recent contribution, taking into consideration
the concepts of competition and composition of capital. It shows that
accumulation is best understood through a class analysis. However, the
capital relation is influential at different levels and it encompasses a broad
range of features of modern society. In this context, this paper proposes a
richer and more encompassing analysis of accumulation.
In the latest issue of Research in Political Economy, Paul Zarembka (2000)
published an important contribution to the analysis of capital accumulation. In
his article, Zarembka shows the potential ambiguities in Marx’s texts, discusses
their implications, and reviews the debates between Lenin, Bukharin,
Luxemburg, Grossman and others.1The full implications of Zarembka’s article
cannot be pursued here. In what follows, I offer a modest contribution to his
analysis, incorporating a more detailed analysis of the composition of capital
and a broader understanding of accumulation.
The composition of capital (including the technical, organic and value
compositions, or TCC, OCC and VCC) is highly important for Marx, and it
plays an important role at several stages in his analysis. For example, the OCC
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
69
Marx’s Capital and Capitalism; Markets in a Socialist Alternative,
Volume 19, pages 69–85.
Copyright © 2001 by Elsevier Science Ltd.
All rights of reproduction in any form reserved.
ISBN: 0-7623-0838-9

is famously the pivot of the transformation problem and the tendency of the
rate of profit to fall,2and it plays a critical role in Marx’s theory of rent,3and
in his analysis of technical change and capital accumulation (see below). In
spite of the importance of this concept, the composition of capital has often
been explained cursorily and understood only superficially and incorrectly in
most of the literature.
This article shows that a clearer understanding of the composition of capital
can help to substantiate and contextualise some of Zarembka’s claims, and
contribute to the development of Marx’s theory of value, exploitation and capital
accumulation. The argument is developed in four sections. The first briefly
reviews the Marxian concepts of capital and exploitation. The second explains
the meaning and significance of the TCC, OCC and VCC in the simplest (static)
case, for heuristic purposes, and in the dynamic context, more relevant for our
purposes. The third analyses in more detail the relationship between capital
accumulation and changes in the technical, organic and value compositions
of capital. The fourth section summarises the argument and draws brief
conclusions.
I. CAPITAL AND EXPLOITATION
I.A. Capital and Wage Labour
For Marx, capital is a social relation between two classes, capitalists and
workers. This relation is established when the means of production are monop-
olised by the capitalists, that employ wage workers in production for profit.
Once this class relation of production is posited, capital exists in and through
things, namely, the means of production, commodities, money and financial
assets employed in the process of valorisation:
Capital is not a thing, any more than money is a thing. In capital, as in money, certain
specific social relations of production between people appear as relations of things to people,
or else certain social relations appear as the natural properties of things in society ...
Capital and wage-labour . .. only express two aspects of the self-same relationship. Money
cannot become capital unless it is exchanged for labour-power . . . Conversely, work can
only be wage-labour when its own material conditions confront it as autonomous powers,
alien property, value existing for itself and maintaining itself, in short as capital . . .
Wage-labour is then a necessary condition for the formation of capital and remains the
essential prerequisite of capitalist production (Capital,1, pp. 1005–1006).4
There is a relationship of mutual implication between capitalism (the mode
of social production), wage labour (the form of social labour), and the
commodity (the typical form of the output):
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
70 ALFREDO SAAD-FILHO

[The] relation between generalised commodity production [GCP] . . . wage labor and
capitalist production is one of reciprocal implication. First . . . when labor becomes wage
labor . . . commodity production is generalised. On the one hand wage labor implies GCP
. . . On the other hand, GCP implies wage labor . . . Marx shows . . . that capitalist production
is commodity production as the general form of production while, at the same time,
emphasizing that it is only on the basis of the capitalist mode of production that all or even
the majority of products of labor assume commodity form . .. Finally, the relation of wage
labor and capital is also one of reciprocal implication for Marx. Capital is a production
relation between the immediate producers and their conditions of production which, separated
from them and passing under the control of non (immediate) producers, dominate them as
capital . .. [T]he rest of the features of capitalism could be seen as the necessary resultants
following from any one of these essentially equivalent central categories (Chattopadhyay,
1994, pp. 17–18).
As a totality engaged in self-expansion through the employment of wage labour,
capital is primarily capital in general. This is the general form of capital.5
Capital in general can be represented by the circuit of industrial capital,
M-C-M’, where M and M’ are sums of money-capital and C represents the
inputs, including labour power and means of production; the difference between
M’ and M is the surplus value.
The circuit of industrial capital represents the essence of capital, valorisation
through the production of commodities by wage labour. However, capital
produces not only surplus value; at the social level, the outcome of the circuit
is the expanded reproduction of capital or, following from the concept of capital,
the renewal of the separation between capitalists and wage workers. For this
reason, Marx claimed that ‘Accumulation of capital is . . . multiplication of the
proletariat’ (Capital,1, p. 764). In other words,
The capitalist process of production . .. seen as a total, connected process, i.e. a process of
reproduction, produces not only commodities, not only surplus-value, but it also produces
and reproduces the capital-relation itself; on the one hand the capitalist, on the other the
wage-labourer (Capital,1, p. 724).
I.B. Exploitation
The capital relation implies that the means of production have been monopo-
lised by a relatively small number of people. In contrast, the majority is forced
to sell their labour power in order to purchase commodities that, as a class,
they have produced previously (see Theories of Surplus Value,3, pp. 490–491).
Therefore, capital is a class relation of exploitation, which allows capitalists to
live off the surplus value extracted from the working class:
Capitalism, and hence capital, requires a lot more by way of the social than private property
and the market . . . What it does depend upon is wage labour, able and willing to produce
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Capital Accumulation and the Composition of Capital 71
71

a surplus for capital. By implication, the social attached to capital takes the form of class
relations . . . Capital and labour confront one another as classes with the capitalist class
monopolising the means of production or access to livelihood through work. Consequently,
workers can only survive by selling their capacity to work for a wage that represents less
in terms of labour time than is performed for the capitalist. The surplus labour performed
over and above that necessary to provide the wage gives rise to what Marx termed exploita-
tion, and provides for the profits of the capitalists (Fine, 2001, p. 29).
For Marx, the defining feature of capitalism is the exploitation of the class of
wage workers by the capitalist class, through the extraction of surplus value.6
The ratio between the surplus value (surplus labour time) and the value of
labour power (necessary labour time) is the rate of exploitation or rate of surplus
value. All else constant, the rate of exploitation can increase for at least three
reasons: if more hours are worked, if the intensity of labour increases, or if the
necessary labour time declines because of productivity growth in the sectors
producing necessities (given the real wage). Marx calls the first two cases the
production of absolute surplus value, while the third produces relative surplus
value (see Capital,1, pp. 430–437, 645–646; Theories of Surplus Value,1,
p. 216; Foley, 1986, pp. 50–54; Lapides, 1998, p. 192). Absolute surplus value
is generally limited, because it is impossible to increase the working day or the
intensity of labour indefinitely, and the workers gradually learn to resist against
these forms of exploitation. In contrast, relative surplus value is more flexible
and harder to resist, because productivity growth can outstrip wage increases
for long periods.
I.C. Competition, Mechanisation, and Class Conflict
Intra-sectoral competition, between firms producing the same use-values,
compels each firm to minimise costs in order to maximise its profit rate. This
type of competition may be associated with different firm strategies. For
example, a longer working day increases the output and may reduce unit costs,
because the transfers from fixed capital are spread across larger batches, and
there is a reduced risk of technical obsolescence (that Marx called moral depre-
ciation) because the machines depreciate physically more quickly. In contrast,
greater labour intensity increases the output, because more simple labour is
performed in the same period, but this does not affect directly the unit value
of the product. Finally, technical progress reduces the simple labour necessary
to produce a unit of the product and, consequently, tends to lower its value:
Production for value and surplus-value involves a constantly operating tendency . . . to reduce
the labour-time needed to produce a commodity, i.e. to reduce the commodity’s value, below
the existing social average at any given time. The pressure to reduce the cost price to its
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
72 ALFREDO SAAD-FILHO

minimum becomes the strongest lever for raising the social productivity of labour, though
this appears here simply as a constant increase in the productivity of capital (Capital,3,
p. 1021).
These technical innovations will be copied or emulated by the rival firms. This
process continually erodes the advantage of the innovating firms, while
preserving the incentives for further technical progress across the economy. At
the level of capital in general, competition and technical change constantly
reduce the value of all goods, including those consumed by the workers. All
else constant, they permit the extraction of relative surplus value:
Capital therefore has an immanent drive, and a constant tendency, towards increasing the
productivity of labour, in order to cheapen commodities and, by cheapening commodities,
to cheapen the worker himself (Capital,1, pp. 436–437).
The most important aspect of intra-sectoral competition is mechanisation, or
the introduction of new technologies and new machines by the innovating firms.
Mechanisation has three principal aspects, two of which were discussed above;
it increases the value-productivity of labour and the profit rate of the innovating
capitals, facilitates the extraction of relative surplus value and, finally, it is a
tool of capitalist control. The Marxian critique of technology has demonstrated
that, underneath their seemingly neutral, scientific and productivist (of use value)
guise, machines are despotic dictators of the rhythm and content of the labour
process.7Therefore, in spite of the perception that competition invariably
increases physical productivity, reduces commodity values and potentially leads
to higher real wages, the relationship between competition and machinery is
complicated by two factors. First, firms do not select the technologies that are
most productive of use values, but those that are most profitable, and these
criteria may lead to distinct outcomes. Second, capitalist attempts to establish
control in the production line and in society may introduce further biases in
the choice of technology, including the adoption of technologies that are not
prima facie more profitable, but that facilitate control (see Levidow & Young,
1981, 1985; Slater, 1980). In sum, conflicts between competing capitals, between
capital and labour on the shopfloor, and between social groups across society,
can influence the choice of technology and the output mix, with consequences
that cannot always be anticipated.
II. COMPOSITION OF CAPITAL AND
ACCUMULATION
A review of the literature shows very diverse understandings of Marx’s concept
of composition of capital. Most writers fail to distinguish between the TCC,
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Capital Accumulation and the Composition of Capital 73
73

OCC and VCC and, when attempting to do so, their views are often distinct
from Marx’s (for a survey of the understandings of the composition of capital,
see Saad-Filho, 1993, 2002, Ch. 6). This section outlines the meaning and
significance of the composition of capital; their implications for the study of
accumulation are explained in Section III.B.
II.A. Static Case
(i) Productivity of Labour and TCC
The productivity of labour is the mass of means of production that can be
processed into final commodities in a given labour time or, alternatively, the
output per hour.8This notion is captured by the technical composition of capital
(TCC, which Marx called earlier the physical composition of capital). The TCC
is the ratio between the mass of material inputs and the living labour necessary
to transform them into the output:
A certain quantity of labour-power, represented by a certain number of workers, is required
to produce a certain volume of products in a day, for example, and this involves putting a
certain definite mass of means of production in motion and consuming them productively
– machines, raw materials etc . . . This proportion constitutes the technical composition of
capital, and is the actual basis of its organic composition (Capital,3, p. 244; see also
Theories of Surplus Value,2, pp. 455–456).
(ii) The OCC
The TCC is the ratio between a heterogeneous bundle of use values and concrete
(even if average) labour; therefore, in most cases it cannot be measured directly,
or contrasted across firms with distinct technologies or producing different
goods. However, the TCC can be assessed in value terms, which defines the
organic composition of capital (OCC). The OCC is a ‘technological composi-
tion’ that synthesises, in value terms, the technical relations of production. More
specifically, the OCC is the value of the means of production (including fixed
and circulating capital) that absorb one hour of living labour (whether paid or
unpaid) in a given firm, industry or economy (see Theories of Surplus Value,
2, pp. 276, 279; Theories of Surplus Value,3, pp. 382, 387).
There is, however, a severe difficulty with this concept. The value of a bundle
of means of production is the product of the values of its components by the
quantities used up. Therefore, it is impossible to tell whether differences
or changes in the OCC are due to differences or changes in the TCC
(i.e. differences or changes in the labour productivity in this industry) or from
differences or changes in the value of the means of production used up (that
reflect the circumstances in other industries). However, for Marx there was no
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
74 ALFREDO SAAD-FILHO

ambiguity. As the OCC is a value-reflex of the TCC, it does not change if the
TCC is constant, whatever may happen to the value of the elements of capital:
if one assumes that the organic composition of capitals is given and likewise the differences
which arise from the differences in their organic composition, then the value ratio can change
although the technological composition remains the same . . . The organic changes and those
brought about by changes of value can have a similar effect on the rate of profit in certain
circumstances. They differ however in the following way. If the latter are not due simply
to fluctuations of market prices and are therefore not temporary, they are invariably caused
by an organic change in the spheres that provide the elements of constant or of variable
capital (Theories of Surplus Value,3, pp. 383–386, various paragraphs; see also Theories
of Surplus Value,1, pp. 415–416; Theories of Surplus Value,2, pp. 376–377).
(iii) The VCC
In order to distinguish clearly between different technologies and the use of
inputs with distinct values, Marx introduces, in Capital, the concept of value
composition of capital (VCC). The VCC is the ratio between the value of the
circulating part of the constant capital (including the depreciation of fixed
capital) and the variable capital (paid labour) necessary to produce the output
(c/v):
The composition of capital is to be understood in a two-fold sense. As value, it is determined
by the proportion in which it is divided into constant capital . . . and variable capital .. .
As material, as it functions in the process of production, all capital is divided into means
of production and living labour-power. This latter composition is determined by the relation
between the mass of the means of production employed on the one hand, and the mass of
labour necessary for their employment on the other. I call the former the value-composition,
the latter the technical composition of capital. There is a close correlation between the two.
To express this, I call the value-composition of capital, in so far as it is determined by its
technical composition and mirrors the changes in the latter, the organic composition of
capital (Capital,1, p. 762.).9
(iv) Implications
Marx’s contrast between the OCC and VCC in the static case allows him to
distinguish clearly between technical and value differences across distinct
production processes. For example, if two firms use the same technology to
produce different products using inputs with distinct value (e.g. copper and
silver jewellery), their TCCs are obviously identical. This implies that their
OCCs are also equal; however, their VCCs are different (in this case, because
silver is more valuable than copper; see Capital,3, pp. 244, 900–901; Theories
of Surplus Value,2, p. 289; Theories of Surplus Value,3, p. 386–387; this
example is inspired by Fine, 1989, pp. 62–63). Similarly, two capitals could
have equal VCCs, even if their TCCs and OCCs were different, depending upon
the input values. These examples show that differences in the value of the
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Capital Accumulation and the Composition of Capital 75
75

constant and variable capital consumed in distinct industries are captured by
the VCC but not the OCC; in contrast, differences in the technologies of
production affect the OCC but they may not be accurately reflected by the
VCC. The concept of OCC is important because it allows the study of technical
differences or changes (see below) in production, regardless of the corre-
sponding value differences or changes that affect primarily the circulation of
capital, while the VCC cannot distinguish between them.
II.B. Dynamic Case
(i) TCC, OCC and VCC
It was shown in Section I.C that technical change is usually introduced in indi-
vidual firms, which raises their TCCs and, consequently, their OCCs and VCCs
(although the three compositions change simultaneously in real time, in logical
terms the TCC changes first, and this shift is reflected by the OCC and, subse-
quently, the VCC). From the point of view of capital in general, its TCC and
OCC tend to rise in every turnover and, all else constant, commodity values
tend to fall. Because of the conflicting forces of competition, including
mechanisation and declining commodity values, the VCC of capital in general
can either rise or fall through time. The outcome depends upon the sectors
affected by technical change, the speed of the diffusion of innovations, the
structure of the systems of provision of commodities, and other factors that can
be analysed only concretely.
In general, however, because of technical progress the values at the beginning
of the circuit (‘earlier values’), at which the inputs are purchased, are higher
than those at which the output is sold (‘later values’). Marx argues that the
OCC reflects the TCC at the initial (higher) values of the component parts of
capital, before the new technologies affect the value of the output, in which
case the social OCC rises in tandem with the social TCC. In contrast, the VCC
reflects the TCC at the final (lower) value of the elements of constant and
variable capital, determined by the modified conditions of production and newly
established in exchange (see Saad-Filho, 1993, 2002, Ch. 6). Therefore, changes
in the social VCC capture the rise in the social TCC and the ensuing fall in
commodity values, including those that have been used as inputs:
This change in the technical composition of capital . .. is reflected in its value-composition
by the increase of the constant constituent of capital at the expense of its variable constituent
. . . However . . . this change in the composition of the value of the capital, provides only
an approximate indication of the change in the composition of its material constituents . . .
The reason is simple: with the increasing productivity of labour, the mass of the means of
production consumed by labour increases, but their value in comparison with their mass
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
76 ALFREDO SAAD-FILHO

diminishes. Their value therefore rises absolutely, but not in proportion to the increase in
their mass (Capital,1, pp. 773–774; see also Capital,3, pp. 317–323).
(ii) Implications
The OCC is distinguished from the VCC only through the comparison between
contrasting situations. If one compares two capitals at the same moment of
time, one would contrast the value of the constant capital productively consumed
per hour of labour (VCC) with the mass of means of production processed in
the same time (TCC and OCC). This case is important theoretically, and it was
through the static comparison of capitals with distinct organic compositions that
Marx developed, in Part 2 of Capital, 3, his transformation of values into prices
of production (see Fine, 1983; Saad-Filho, 1997, 2002, Ch. 7).
In a dynamic context, both the OCC and VCC of an individual capital (or
capital in general) undergoing technical change can be calculated. These compo-
sitions can diverge because the OCC is an ex ante evaluation of the (fixed and
circulating) constant capital technically required per hour of (paid and unpaid)
labour, while the VCC is the ex post ratio between the new value of the
(circulating) constant and the variable capital spent in the last phase of
production. Thus, the OCC is measured at the time of production, while the
VCC is determined in exchange and calculated on the basis of the values newly
established by the currently predominant technologies. It was in this context
that Marx presented his law of the tendency of the rate of profit to fall, in Part
3 of Capital 3 (see Fine, 1989, Ch. 10; 1992).
More importantly, the distinction between TCC, OCC and VCC helps to
illuminate the potential implications of capital accumulation. Technical change
raises the TCC, the OCC and total input values (because the output increases).
However, the VCC, unit output values and future input prices tend to fall. How
the actual process of adjustment happens – especially for large blocs of fixed
capital – is crucial, because the sudden devaluation of large masses of capital
can lead to financial upheaval and crises (see Perelman, 1993, 1999).
III. ACCUMULATION AND THE COMPOSITION
OF CAPITAL
III.A. Zarembka’s Analysis
Zarembka argues that Marx’s definition of accumulation of capital is imprecise
and potentially misleading (see p. 185):10
Marx first seems to define accumulation to necessarily include an increase in the number
of workers under the domination of capital, as both constant capital and variable capital
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Capital Accumulation and the Composition of Capital 77
77

increase. Then, accumulation seems to include a case where the number of workers stays
the same or even declines. But then another passage suggests that accumulation could take
place even without a change in constant capital. So, what is accumulation of capital? Could
accumulation of capital . . . be simply c+v increasing with the proportions between cand
vunimportant for the definition? . . . Most importantly, is it consistent with the very concept
of “capital” in Marx? These questions are unresolved in Marx’s writing and became an
ambiguity embedded in twentieth-century Marxism (pp. 196–197).
The concept of capital, for Zarembka, is that explained in Section I, i.e. the
relationship between the capitalist class and the class of wage-laborers (see
pp. 184–185). In the light of this (class, rather than merely technical, productivist
or monetary) concept of capital, Zarembka argues (p. 185) that accumulation
has been confused in the Marxian tradition with issues surrounding the organic
composition of capital:
It has been a virtual dogma within Marxism for a century that, as accumulation takes place,
the ratio of constant capital to variable capital rises, i.e. the “organic composition of capital”
c/v rises . . . Yet [for Marx] . . . the opposite can occur . . . constant capital may not rise
at all – or less than variable capital – as accumulation occurs (pp. 195–196).
For Zarembka, the debate about the relationship between capital accumulation
and the growth in constant capital is misplaced and potentially misleading:
the essential factor of accumulation of capital is the increase in wage-labor, not an
increase in constant capital. Note [Marx’s] reference to either “more capitalists” or “larger
capitalists”, but when the issue comes to labor it is simply “more wage-workers” – only
the number is important (p. 192).
Given his focus upon the class relation between capitalists and workers, rather
than the rising quantity of machines, output or money-capital, Zarembka offers
the following definition:
“accumulation of capital is increase of proletarian labor with its associated constant capital”
or, in more modern language, “accumulation of capital is increase of wage-labor with
its associated constant capital”. Accumulation will generally include additional constant
capital. If the process of accumulation of capital is related to the production of relative
surplus value (i.e. technological improvements), as is typically the case, it should be
discussed as such. Whether the organic composition of capital also increases is not directly
germane to the issue if and how much capital is accumulating . . . if there is no increase
in wage-labor, there is no accumulation . . . As to increased wage-labor, it can come from:
(1) longer work hours,
(2) population increase, and
(3) new proletarianizations concomitant with creation of home markets and markets abroad,
including drawing in of new sections of the population such as women and children . . .
The definition of accumulation offered here leads to the deepest issues of
Marxist thought, the relations and struggles between and among social classes
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
78 ALFREDO SAAD-FILHO

. . . It focuses the question to the class issues and away from production or
from productivity (pp. 223–224).
III.B. Accumulation, Class, and the Composition of Capital
The class interpretation of capital, explained in section I.A, is the premise of
Zarembka’s definition of accumulation. This section shows that the analysis
of exploitation and competition, in Sections I.B and I.C, and the distinction
between TCC, OCC and VCC, explained in Section II, can contribute to a more
precise analysis, that enriches our understanding of capital accumulation and
contextualises Zarembka’s approach. The close relationship between the compo-
sition of capital and accumulation is not surprising, for Marx analyses both
simultaneously, most clearly in chapter 25 of Capital,1.
It was shown in Section I.A that capital is a class relation that appears through
the transformation of goods, services, machines, money, financial assets and
labour power into means of valorisation, M-C-M’. The distinction between
the essence of capital and its forms of appearance opens the possibility of
ambiguity in the definition of accumulation; for example, Zarembka shows
that accumulation has been understood in different ways in the Marxian
literature, with potentially important consequences.11 Definitions of accumula-
tion can focus upon the growth of output, money capital advanced, labour
productivity or means of production in use (either more or better machines),
or upon the growth of labour productivity or employment (Zarembka’s own
definition). Moreover, these distinct understandings of accumulation can refer
to individual capitals or to capital as a whole, depending upon the level of
analysis.
Whatever its definition, accumulation is closely related with competition and
crisis. Competition forces capitals to maximise profits, which provide the
resources for investment and growth; however, competitive pressures give rise
to crisis tendencies that can lead to disastrous outcomes, especially bankruptcies,
economic depression and high unemployment (see Clarke, 1994). These rela-
tions explain Marx’s claim that ‘the employment of surplus-value as capital,
or its reconversion into capital, is called accumulation of capital’ (Capital,1,
p. 725). Reconversion of surplus value into capital, or expanded reproduction,
can occur under two circumstances, extended expanded reproduction, or
intensified expanded reproduction (Fine & Harris, 1979, pp. 112–115). The
former is associated with the replication of current technologies, or output
growth with constant productivity, while the latter explains output growth
through productivity increase.
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Capital Accumulation and the Composition of Capital 79
79

(1) Extended Expanded Reproduction:
(1a) Growth in the quantity of means of production: For the individual capital,
increasing the quantity of means of production (machines) replicating the same
technology increases the number of employees, total output and the profit mass,
while the TCC, OCC and VCC, the productivity of labour and the rate of profit
remain constant. The outcome is identical for capital as a whole, and commodity
values remain unchanged.
(1b) Growth in the quantity of simple labour employed: Three possibilities exist:
ii(i) Employment growth due to an increase in the quantity of means of produc-
tion, as in scenario 1a.
i(ii) Employment growth due to an increase in the number of hours worked
(absolute surplus value); in this case, for the individual capital, the TCC,
OCC and labour productivity remain constant, but the VCC declines and
the output and the rate of profit increase because of the decline in average
fixed costs. The same is the case for capital as a whole; values may decline
due to the lower input values.
(iii) Employment growth due to the substitution of labour for machines: for
the individual capital, the TCC, OCC and VCC decline, with uncertain
consequences for productivity, output and profit rate. It is the same for
capital as a whole, and values may or may not change. The general rate
of profit rises (more living labour produces more surplus value).
(2) Intensified Expanded Reproduction:
(2a) Rising productivity of labour, given the stock of machines (higher intensity
of labour or better training of the workforce): the TCC and OCC remain
constant but the VCC declines and labour productivity increase (given the wage
rate) because more simple labour is condensed in each hour of concrete labour.
Therefore, the output and the profit rate of the individual capital increase
(absolute surplus value). For capital as a whole, commodity values decline
because ‘simple’ labour becomes more productive.
(2b) Rising efficiency of the means of production (better machines with given
employment of labour): the TCC, OCC and VCC, labour productivity, output
and the profit rate increase, with falling unit costs (unless new machines are
introduced because of control rather than profitability, see Section II.C). For
capital as a whole, the TCC and OCC increase, with uncertain effects on the
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
80 ALFREDO SAAD-FILHO

VCC and the rate of profit because values tend to decline (relative surplus
value).12 This is the context of Marx’s analysis of the tendency of the rate of
profit to fall, the counter-tendencies, and the possibility of crisis associated with
them (Fine, 1992; see also Research in Political Economy, 17, 1999, 18, 2000).
Scenarios 1a and 2b are associated with more or better machines, and 1b and
2a with more labour. All scenarios (except, possibly, 1b(iii)) lead to output
growth, and all except, possibly, 1b(iii) and 2b, require a greater advance of
money capital. These potentially very different scenarios purporting to represent
capital accumulation substantiate Zarembka’s claim that, for reasons of clarity
and comprehensiveness, analysis of accumulation should transcend the appear-
ances and focus upon the essence of capital. He is also right to focus upon the
class relations underlying capital accumulation.
However, Zarembka’s approach is limited in two important ways. First,
Zarembka defines accumulation by the quantity of labour employed, with no
further qualification, potentially implying that the number of workers is the
decisive variable. This is insufficient even in its own terms for, as 1b(ii) and
2a show, the same number of workers can perform more simple labour through
longer hours, greater exertion or better training. More generally, Zarembka’s
analysis cannot readily distinguish between scenarios 1b and 2a, and it implicitly
rejects 1a and 2b as being unrelated to accumulation.
By the same token, it is difficult to accept the claim that the displacement
of machines by workers necessarily implies capital accumulation. For example,
binding foreign exchange or environmental constraints can force this substitu-
tion, potentially leading to a simultaneous reduction in output, the TCC,
OCC, VCC and labour productivity, which is hardly conducive to capital
accumulation. Conversely, technological unemployment, because cheaper or
better machines are available, may lead to higher productivity, TCC and OCC,
output growth and lower commodity values (relative surplus value). However,
for Zarembka this does not imply capital accumulation unless employment also
rises, although the latter plays only a minor role in this form of intensified
expanded reproduction.
The potentially conflicting definitions of capital accumulation identified above
are symptomatic of three broader difficulties. First, the capitalist use of machines
depends upon two related but distinct variables, profitability and control; it does
not depend directly upon either productivity or the level of employment.
Therefore, not only accumulation (however defined) and productivity can move
in opposite directions, but the imperatives of competition, profitability and
workplace control may come into conflict, leading to uncertain outcomes in
terms of mechanisation and accumulation. These conflicting imperatives may
have severe consequences for production, profitability and accumulation.
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Capital Accumulation and the Composition of Capital 81
81

Second, although in most countries (primitive) accumulation generally
involves the absorption of increasing numbers of workers into the formal wage
labour markets, this is not necessarily the case. Across the world, the expanded
reproduction of capital can be based upon the extraction of absolute or relative
surplus value, or the abuse of natural resources for consumption or export, in
which case the size of the workforce is only one aspect of accumulation. In
developed countries, where wage labour prevails, the size of the workforce and
the level of wages are not the determining variable; they are determined by the
rhythm of accumulation, rather than vice-versa. For example, Marx argues that:
It is [the] movements of the accumulation of capital which are reflected as relative
movements of the mass of exploitable labour-power, and therefore seem produced by the
latter’s own independent movement. To put it mathematically: the rate of accumulation is
the independent, not the dependent variable; the rate of wages is the dependent, not the
independent variable (Capital 1, p. 770).
Similarly, recent experience across the world shows that sustained growth and
the expansion of the economic and political power of capital are compatible
with increasing exploitation of the workers and rising unemployment.
Third, accumulation synthesizes the contradictions of capital. Capitalist
production is associated with competition, economic growth, sectoral imbal-
ances, and crisis. At this level of analysis, too, the number of employed workers
is an important but not decisive variable. As long as wage labour is the form
of social labour, and capital rules the reproduction of the labour force, including
the state, the legal process, police and the armed forces, the credit system,
education, training and immigration rules, the size of the labour force is an
important but not binding constraint.
IV. CONCLUSION
This article contributes to the development of Zarembka’s (2000) concept of
capital accumulation, in the light of a class analysis of capital and the distinction
between the technical, organic and value compositions of capital.
Zarembka’s definition of accumulation as the development of the relationship
between capitalists and wage workers, or the increase in the number of wage
working hours performed, offers an important insight into the contradictory
world of capital. For example, it shows that capital accumulation is compatible
with rising or even falling productivity, the use or withdrawal of machines, and
any manner of changes in input and output values.
Development of this approach through the analysis of competition and
the composition of capital shows, first, that accumulation synthesizes the
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
82 ALFREDO SAAD-FILHO

contradictions of capital both in production and exchange. Second, there are
potential conflicts between the imperatives of profitability and control, which
are expressed through competition and mechanisation. Third, shifts in
commodity values, due to mechanisation, can lead to crisis. The approach
developed in this article implies that economic crises can be due to complex
combinations of factors, including technical change, disproportions due
to competition and mechanisation, financial upheaval, and class conflict.
None of these potential causes of crisis can explain empirical phenomena in
isolation but, in the context of a class analysis, they can help to illuminate
empirical developments more powerfully and insightfully than conventional
interpretations.
NOTES
1. For a more detailed analysis of these debates, see Zarembka (2001).
2. See, respectively, Capital, 3, Parts 1–2, Fine (1983) and Saad-Filho (1993, 1997,
2002, Chs. 6–7), and Capital, 3, Part 3 and Fine (1992). The composition of capital is
analysed by Fine and Harris (1979, Ch. 4), Fine (1989, Ch. 10; 1990), Meacci (1992)
and Weeks (1981, Ch. 8). The tendency of the rate of profit to fall has been the subject
of debates in Research in Political Economy, 17 (1999) and 18 (2000).
3. See Capital, 3, Part 6, Theories of Surplus Value, 2, Chs. 1–14 and Fine (1989,
Ch. 13).
4. Chattopadhyay (1994, p. 18) rightly argues that ‘Marx’s starting point in the
treatment of capital is conceiving capital as a social totality, capital representing a class
opposed not so much to the individual laborers as to the wage laborers as a class’.
5. See Grundrisse, pp. 310, 449, 852.
6. ‘To Marx . . . the essence of capitalist property is the control of the productive
process and therefore the control over laborers. Forced labor rather than low wages,
alienation of labor rather than alienation of the product of labor are, according to Marx,
the essence of capitalist exploitation’ (Medio, 1977, p. 384).
7. See Attewell (1984), Bowles and Gintis (1977), Braverman (1974), Brighton
Labour Process Group (1977), Cleaver (1979, 1992), Lebowitz (1992), Levidow and
Young (1981, 1985), Marglin (1974), Postone (1993), Slater (1980), Sohn-Rethel (1978)
and Spencer (2000).
8. See Capital, 1, pp. 137, 431, 773, 959.
9. Alternatively, ‘The organic composition of capital is the name we give to its value
composition, in so far as this is determined by its technical composition and reflects it’
(Capital, 3, p. 245); see also Harvey (1999, p. 126) and Weeks (1981, pp. 197–201).
10. In this section, page numbers refer to Zarembka (2000) unless stated otherwise.
11. For detailed Marxian analyses of accumulation, see Fine (1989, Ch. 5), Harvey
(1999, Ch. 6) and Weeks (1981, Ch. 8).
12. In this case, ‘The process of accumulation involves the initiation of the circuit
of capital upon the basis of one set of values, and the generation of a new set of values
that confronts capitalists at the end of the circuit’ (Weeks 1981, p. 194).
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Capital Accumulation and the Composition of Capital 83
83

ACKNOWLEDGMENTS
I am grateful to Andrew Brown and Alejandro Ramos-Martínez for their helpful
comments. I am, however, responsible for the remaining errors and omissions.
REFERENCES
Aglietta, M. (1979). A Theory of Capitalist Regulation, the U.S. Experience. London: New Left
Books.
Attewell, P. A. (1984). Radical Political Economy, A Sociology of Knowledge Analysis. New
Brunswick, N.J.: Rutgers University Press.
Bowles, S., & Gintis, H. (1977). The Marxian Theory of Value and Heterogeneous Labour, Critique
and Reformulation. Cambridge Journal of Economics, 1(2), 173–192.
Braverman, H. (1974). Labour and Monopoly Capital. New York: Monthly Review Press.
Brighton Labour Process Group (1977). The Capitalist Labour Process. Capital & Class, 1, 3–26.
Chattopadhyay, P. (1994). The Marxian Concept of Capital and the Soviet Experience: Essay in
the Critique of Political Economy. Westport, Conn.: Praeger.
Clarke, S. (1994). Marx’s Theory of Crisis. London: Macmillan.
Cleaver, H. (1979). Reading ‘Capital’ Politically. Brighton, The Harvester Press.
Cleaver, H. (1992). The Inversion of Class Perspective in Marxian Theory: From Valorisation to
Self-Valorisation. In: W. Bonefeld, R. Gunn & K. Psychopedis (Eds), Open Marxism.
London: Pluto Press.
Fine, B. (1983). A Dissenting Note on the Transformation Problem. Economy & Society, 12(4),
520–525.
Fine, B. (1989). Marx’s Capital (3rd ed.). Basingstoke, Macmillan.
Fine, B. (1990). On the Composition of Capital, A Comment on Groll and Orzech. History of
Political Economy, 22(1), 149–155.
Fine, B. (1992). On the Falling Rate of Profit. In: G. A. Caravale (Ed.), Marx and Modern Economic
Analysis. Aldershot: Edward Elgar.
Fine, B. (2001). Social Capital versus Social Theory. London: Routledge.
Fine, B., & Harris, L. (1979). Rereading Capital. London: Macmillan.
Foley, D. (1986). Understanding Capital, Marx’s Economic Theory. Cambridge, Mass.: Harvard
University Press.
Harvey, D. (1999). The Limits to Capital. London: Verso.
Lapides, K. (1998). Marx’s Wage Theory in Historical Perspective: Its Origins, Development and
Interpretation. Westport, Conn.: Praeger.
Lebowitz, M. (1992). Beyond Capital, Marx’s Political Economy of the Working Class. London:
Macmillan.
Levidow, L., & Young, B. (1981, 1985). Science, Technology and the Labour Process, Marxist
Studies, 2 Vols. London: Free Association Books.
Marglin, S. (1974). What Do Bosses Do? Review of Radical Political Economics, 6(2), 60–112.
Marx, K. (1978a, 1969, 1972). Theories of Surplus Value, 3 Vols. London: Lawrence and Wishart.
Marx, K. (1981a). Grundrisse. Harmondsworth: Penguin.
Marx, K. (1976, 1978b, 1981b). Capital, 3 Vols. Harmondsworth: Penguin.
Meacci, F. (1992). The Organic Composition of Capital and the Falling Rate of Profit. In:
G. A. Caravale (Ed.), Marx and Modern Economic Analysis. Aldershot: Edward Elgar.
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
84 ALFREDO SAAD-FILHO

Medio, A. (1977). Neoclassicals, Neo-Ricardians, and Marx. In: J. G. Schwartz (Ed.), The Subtle
Anatomy of Capitalism. Santa Monica: Goodyear.
Nell, E. J. (1992). Transformational Growth and Effective Demand. New York: New York
University Press.
Perelman, M. (1993). The Qualitative Side of Marx’s Value Theory. Rethinking Marxism, 6(1),
82–95.
Perelman, M. (1999). Marx, Devalorisation, and the Theory of Value. Cambridge Journal of
Economics, 23(6), 719–728.
Postone, M. (1993). Time, Labour and Social Domination, A Re-examination of Marx’s Critical
Theory. Cambridge: Cambridge University Press.
Saad-Filho, A. (1993). A Note on Marx’s Analysis of the Composition of Capital. Capital & Class
50, 127–146.
Saad-Filho, A. (1997). An Alternative Reading of the Transformation of Values into Prices of
Production. Capital and Class, 63, 115–136.
Saad-Filho, A. (2002). The Value of Marx: Political Economy for Contemporary Capitalism.
London: Routledge.
Slater, P. (Ed) (1980). Outlines of a Critique of Technology. Atlantic Highlands: Humanities Press.
Sohn-Rethel, A. (1978). Intellectual and Manual Labour: a Critique of Epistemology. London:
Macmillan.
Spencer, D. (2000). Braverman and the Contribution of Labour Process Analysis to the Critique
of Capitalist Production – Twenty Five Years On. Work, Employment and Society, 14(2),
223–243.
Weeks, J. (1981). Capital and Exploitation. Princeton: Princeton University Press.
Zarembka, P. (2000). Accumulation of Capital, its Definition: A Century after Lenin and Luxemburg.
Research in Political Economy, 18, 183–225.
Zarembka, P. (2001). Rosa Luxemburg’s Accumulation of Capital: Critics Try to Bury the Message.
Current Perspectives in Social Theory, 21 (forthcoming).
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Capital Accumulation and the Composition of Capital 85
85

1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
86 ALFREDO SAAD-FILHO