1991 article, originally published in Cambridge Journal of Economics 15(1): 79–93.
(For its abstract see the Abstracts of all chapters, p. 9.)
Contents
Introduction
1 Accumulation of capital
2 Tendencies of capital interaction and capital stratification
2.1 Inter- and intra-branch interaction
2.2 Stratification
2.3 Tendencies versus trends
3 Stratification and the money expression of labour and abstract labour
4 Tendencies of accumulation of capital
4.1 The tendency towards the over-accumulation of capital
4.2 The tendency for the value composition of capital to rise and the tendency of the rate of profit to fall
5 The contradiction of the TRPF
6 The conditions of existence of the contradiction of the TRPF: stratification and devalorisation
7 The TRPF and devaluation of capital
8 The TRPF and the restructuring and centralisation of capital
Summary and conclusions
References
Annotation (of 2023). This article sets out a foundation of the theory of “the tendency of the rate of profit to fall” (TRPF) in the “stratification of capital”. As remarked in the article’s footnote 3, the upshot of the article is, first, that the tendency at hand not only interacts with other tendencies of accumulation (§4), but, secondly, that the expressions of this tendency are likely to be manifest in a cyclical manner (§7–§8). Furthermore, as remarked at the end of §6, capital stratification by itself does not prove the existence of the TRPF, rather, the stratification of capital is generally the typical form of existence of capital. See also the six conclusions of the article.
Introduction1
The rate of profit tends to fall. In 1871 W.S. Jevons comments on this thesis: “There are sufficient statistical facts, too, to confirm this conclusion historically. The only question that can arise is as to the actual cause of this tendency” (1871, p. 246). Indeed the theory of the tendency of the rate of profit to fall (TRPF) was a substantial part of mainstream economic theorising from Adam Smith until the end of the nineteenth century. In more recent times, however, theory of it has been confined almost entirely to marxist or marxist-inspired work, where the tendency is related to the increase in the composition of capital associated with labour-expelling technical change.2 However, even here the issue remains controversial, due partly to the fact that there is little agreement as to the appropriate level of abstraction. Some authors interpret the tendency as an empirical trend of either short-run cyclical or long-run secular development. The view proposed in this article, on the other hand, is that a tendency is a concept belonging to a specific level of abstraction, and so is quite different from the concept of an empirical trend.3
Another important controversy relates to the question of how technical change could ever produce a fall in the rate of profit, when new techniques are presumably only introduced in order to secure increases in the rate of profit. Thus, it is argued, rate of profit decreasing technical change will just not come about: if the rate of profit is to decrease, it must be for reasons other than changes in the composition of capital brought about by technical change (such as wage rate increase). Therefore, many authors (drawing in this respect on the seminal article by Okishio 1961) regard the TRPF as being insufficiently grounded; it lacks (as Roemer 1979 expresses it) a microeconomic foundation. This question is the main issue of this article.
In §5 it is indicated that the TRPF is based on contradictory forces in the economy (indeed new techniques are, in general, only introduced if these secure a higher rate of profit). Nevertheless, contradictory forces may coexist; the question is how these are reconciled. It will be argued in §6 that, besides the fact that authors such as Okishio and Roemer cast their analysis in one-sided physical (use-value) terms, their analysis is inadequate in that it is comparative static. Once the theory is cast in dynamic terms, conditions of existence for the TRPF (or, appropriate “microeconomic foundations”) can indeed be provided, and the analysis of the “Okishians” reduces to a special case. However, whilst these conditions of existence may resolve these contradictory forces, they do not dissolve them. Their expression then is in the devaluation (§7) and the restructuring and centralisation of capital (§8). Before this, §1 introduces the accumulation of capital, and §2–§3 provide the elementary dynamic framework of the interaction of capitals. In §4 the tendency for the composition of capital to rise, and the derived tendency of the rate of profit to fall, is presented as merely one form of existence of the accumulation of capital; the other form being the tendency to over-accumulation of capital, and the derived labour-shortage profit squeeze.
1 Accumulation of capital
The inherent logic of capitalist production is valorisation, the expansion of value; more specifically production is geared towards continual increase of profit. This is achieved firstly by an increase in control over the labour process by capital, and secondly by accumulation of capital. The inherent limits of these are overcome by technical change. (Value is conceived of as a category which finds concrete expression only in money; it should not be taken to be a pre-market concept such as “labour embodied” – see Reuten 1988a and Reuten and Williams 1989, ch. 1.)
Firstly, the fact that production is geared to the increase in profits implies that capital is continually driven to increase control, with the effect of decreasing costs per unit of physical output and increasing the rate of profit. There are clearly limits to such cost reduction. For a given length of the working day and technique of production, the intensity of labour cannot be increased indefinitely. Thus, secondly, capital is enlarged, to be reproduced on an extended scale, via investment of profit, thereby extending the valorisation of capital to the accumulation of capital.
As valorisation is limited by the possible increase in the intensity of labour, this also limits the extent of accumulation of capital. The investment of capital in new techniques of production overcomes these limits. A new technique may both itself reduce unit costs and create the possibility for new organisational techniques to increase the intensity of labour, and thus decrease unit costs even further. With its introduction not only (potential) profits but also the rate of (potential) profit on the newly accumulated capital tends to increase.
2 Tendencies of capital interaction and capital stratification
2.1 Inter- and intra-branch interaction
The interaction of capitals is first determined at the level of inter-branch interaction. Capitalist production is indifferent to the particular use-values produced. Valorisation, the driving force of capital, determines that capital valorised and validated in the one branch may flow – via the mediation of money capital – to be accumulated in another, in pursuit of a higher rate of profit. Inter-branch interaction and accumulation of capital thence establish a tendency of equalisation of average rates of profit between branches.
Secondly, the interaction of capitals is determined at the intra-branch level. This interaction in product markets is induced by the compulsion continuously to realise the output produced. In connection with the generally temporary character of sales and purchase contracts (in general contractual sales above the average market price in one period would have the effect of repelling buyers in the next), it establishes a tendency towards uniform prices in a market. This tendency also applies to labour-power – it is again predicated on the temporary character of sales and purchase contracts (for labour-power) – so that the competition among capitals for labour establishes a tendency for wages, and the intensity of similar labour, to become uniform across capitals. The articulation of these tendencies ensures that the profits of any one capital come to depend on the technique of production adopted. The interaction of capitals therefore reinforces and reproduces more concretely the compulsion to accumulate capital in new techniques of production (§1).
2.2 Stratification
Accumulation of capital in new and cost reducing techniques of production applies to inter- as well as intra-branch investment of capital. In both cases the initiating capital secures an extra profit. The consequent threat of price competition and the necessity for continuous valorisation compels competitors to follow suit. However, each capital is burdened with the fixed costs of its already accumulated capital and will thus only scrap old plants when a new technique offers net profits (taking into account the capital of the old plant foregone) greater than the profits on its existing plant. Since, therefore, plants embodying new technology will in general not be immediately adopted by all capitals, each branch of production tends to be composed of a stratification of capitals dated according to cost of production, and associated rate of profit differences.
2.3 Tendencies versus trends
The concept of a tendency should be distinguished from that of an empirical trend. Though tendencies cannot be taken to be empirical statements in the sense that their effects can be directly observed, they do affect the concrete. To what extent tendencies are actualised (for example, the extent to which rates of profit are indeed equalised, or to what extent the TRPF results in an actual fall of the rate of profit) can never be established at the level of abstraction at which they have been derived (see also Weeks 1981, p. 205; and Cutler, Hindess, Hirst and Hussain 1977, vol. 1, ch. 6.). But before the theory can be confronted with the empirical, the interconnection of tendencies (as well as their articulation with contingencies) has to be theorised.
3 Stratification and the money expression of labour and abstract labour
The interplay of valorisation, accumulation and the tendencies of capital interaction (together with the credit system – see Reuten 1988b) constitutes the value of labour, of commodities and of labour-power – quantitatively expressed as prices in terms of money.4
The capitalist production process is a two-fold process of production of use-value and value (valorisation). Market prices have to be anticipated, thus the labour process is pre-commensurated in terms of the anticipated output price. Because of the articulation of the tendencies of stratification and uniform prices, the anticipated value, m, of the labour-time in each unit of capital (plant) i generally differs. If the anticipated value of the average labour used up in some branch (k) is mk then, in general, because of the stratification of capital, anticipated values will differ within each branch
mi ≠ mj (1)
(where mi is the anticipated money expression of labour in plant i, measured in terms of some standard of money per unit of labour-time). Across branches, anticipated values will in general also differ
mk ≠ ml (2)
Thus these anticipated values are pre-market entities. It is only in the market that the commodity is actualised as an entity of double form (use-value and money). Actual prices are first determined, not by the labour used up in some particular plant according to the technique adopted in that plant, but as a recursive process by the labour used up in that plant in comparison to that required by the socially necessary technique (conceived for the present as the average technique). In this particular sense the price of a commodity is determined by socially necessary labour. The anticipated value mi may thus diverge from the actual validation (“realisation”)
mi ≷ m´i ; mk ≷ m´k (3)
where m´i and m´k are the realised money expressions of labour. Then m´ili is the expression for abstract labour of plant i. Summing the labour realised in each plant (or branch), abstract labour is
Σ m´ili ≡ m´l ≡ Y´; ml ≷ m´l (4)
(where li and l are, respectively, the private and social aggregate labour expended, and Y´ is the social aggregate value-added).5
4 Tendencies of accumulation of capital
4.1 The tendency towards the over-accumulation of capital
With increasing control over the labour-process and the intensity of labour, via the introduction of new techniques of production, the rate of surplus-value tends to rise. This is reflected in a tendency for the rate of accumulation to increase, an ultimate condition of which is a relatively abundant labour-force. Labour abundance may prevent increasing labour intensity being reflected in wage increases and facilitates the reflection in wage decreases of price decreases resulting from any decrease in unit costs. Ultimately, however, the accumulation of an increasing mass of surplus-value must deplete the reserve of labour, which, via an upward pressure on wages, counteracts the increase in the rate of surplus-value, eventually causing it to decrease, leading to a decrease in the rate of accumulation of capital.6 Thus accumulation tends to take the form of over-accumulation of capital: it is extended up to the point where capital becomes abundant relative to labour, because of excessive valorisation.
One expression of this tendency towards the over-accumulation of capital is its effect on labour costs and production, thus it takes on the form of a labour-shortage profit squeeze.7 The resulting interactions may give rise to a cyclical pattern of profits, accumulation and unemployment. (Note that the term cycle is used in the sense of fluctuation in general, without any specification as to duration.)8
4.2 The tendency for the value composition of capital to rise and the tendency of the rate of profit to fall
The accumulation of capital in new techniques of production overcomes the constraint to increasing valorisation of capital imposed by the inherent limits to increasing labour intensity. The compulsion to introduce new techniques of production – reproduced by the interaction of capitals – is expressed in the continual change of the process of production, typically in the form of increase in labour productivity (§1–§2). This implies that in use-value terms, a unit of labour tends to work up an increasing mass of means of production (raw materials and depreciating fixed means of production), so that the technical composition of capital (TCC) tends to increase. It also implies, that at the point in time when a TCC-increasing technique is introduced, the composition of capital in terms of the prices at that point in time increases (that is the instantaneous value composition of capital (VCC) – the ratio of the value of the means of production to the value of labour-power – increases).9 Therefore, an increasing share of any unit of profit accumulated tends to be invested in means of production, and a decreasing share in the wage fund. There is thus a tendency towards relative expulsion of labour.
The tendency for the TCC to increase thus counteracts the tendency towards the over-accumulation of capital; it retards, or even prevents, the eventual scarcity of labour and the associated upward pressure on wages. However, both labour-using accumulation and its negation in relative labour-expelling accumulation derive independently from the accumulation of capital and are therefore to be theorised first independently.10
The rate of profit r is the ratio of profit (value-added, m´l, minus wages wl) to capital laid out (K+wl):
r = [(m´ – w)l] / [K + wl] = [(m´ – w)/w] / [K/(wl) + 1] (5)
(where K is capital invested in fixed and circulating means of production; w is the average money wage rate; m´ is the money expression of social labour; and l is labour (measured in time); K/(wl) is then the value composition of capital). (Note that the components of capital are measured in disequilibrium rather than in equilibrium prices.)
The tendency for the VCC to rise, generates the tendency of the rate of profit to fall (TRPF), leading to a tendency for the rate of accumulation of capital to decrease. Formally a tendential fall in the rate of profit may be offset by a sufficient rise in the rate of surplus-value (m´-w)/w. However, m´ and w are macroeconomic categories and their determination does not directly derive from changes in the composition of capital. Firstly, as for w, in the absence of inflation, an increase in the VCC and the social use-value productivity of labour tends (in the absence of quantity rationing – see §6 below) to be accompanied by decreasing commodity prices. But this does not immediately lead to changes in the wage rate, though the purchasing power of the wage is affected by it.11 Changes in the wage rate are determined rather by the articulation of changes in the intensity and organisation of the labour process, and unemployment (§4.1).
Secondly, concerning m´, because there is a tendency for the effect of increases in labour productivity on m´ to be offset by concomitant price falls (that is in the absence of inflation), an increase in the social average use value productivity of labour need have no effect on social income m´l.12 However, when m´ is constant, value-added (m´l≡
Thus because the TRPF does not by itself immediately affect the macroeconomic categories m´ and w, at this level of abstraction, the rate of surplus-value may be kept constant.13 It does however immediately affect the micro and branch-level categories m´i and m´k, and the concomitant rates of surplus-value (see §6).
5 The contradiction of the TRPF
The TRPF contradicts the logic of accumulation of capital as being geared towards increase in profit and the rate of profit (§1). This has led authors such as Okishio (1961), Himmelweit (1974) and Roemer (1979) to reject the TRPF.14 The critique of these authors is usually considered to be the most telling, and their critique is relevant to the conceptualisation in this article so far. They would agree that: (1) the empirical discussion of whether the composition of capital is rising and the rate of profit falling does not bear upon the theoretical argument (“The empirical investigations, then, are certainly necessary, but they cannot provide refutation of a theory.” – Roemer, 1979, p. 380); (2) an eventual refutation of the theory does not depend on a change in the rate of surplus-value (consequently the part of their argument relevant to this is couched in terms of the maximum rate of profit, with wages approaching zero).
Basicly their critique is that the TRPF lacks a microeconomic foundation. As capitals are impelled to increase profits and the rate of profit, they will only introduce new techniques that are cost-reducing, so, the argument goes, the average rate of profit cannot decrease. Typically, the average rate of profit will increase in terms of both old and new equilibrium prices.15 As it stands all this is quite correct.16 The crucial question is, however, to what extent their comparative static equilibrium account in Sraffian terms adequately deals with the dynamic problematic of the TRPF (cf. also Fine 1982, pp. 112–15, and Weeks 1982). Within the “Okishian” model, all new least-cost techniques are adopted by all capitals. Whilst this “procedure” may seem intuitively adequate when all capital is conceived of as circulating capital (as with Okishio and Himmelweit), it is not when fixed capital is taken into account (as with Roemer, challenged to do so by Shaikh 1978). Capitals will not in general adopt a new technique when its expected increase in revenue and the rate of profit does not compensate for the early obsolescence of the fixed capital of the old technique (see Section 6). But Roemer evades the dynamic problematic of the TRPF because of his odd conceptualisation of fixed capital. In one of his models, he assumes that all fixed capital lasts forever, and proposes that “if the rate of profit can be shown to rise as a consequence of technical innovation in a model when fixed capital lasts forever, a fortiori it should rise when fixed capital wears out, …” (1979, p. 385). The static conception is shown here very clearly: the fixed capital does not need to be replaced. All capitals within a sector of production are homogeneous, so the problem of eventual devaluation of fixed capital (when prices decrease) which hits different capitals unevenly according to their state of amortisation, is defined away.17 The same goes for his Von Neumann type model where he assumes that only those processes are operating “which produce a maximum profit rate”.
However, this critique of the “Okishians” does not preclude the possibility that they are correct in arguing that the “micro-foundations” of the TRPF are indeed inadequate, or even lacking altogether.
6 The conditions of existence of the contradiction of the TRPF: stratification and devalorisation
The contradiction of the tendency of the overall rate of profit to fall and the tendential increase in profit and the rate of profit along with the accumulation of capital in new plants is resolved in the concretisation of the concept of capital stratification. It was argued above (§2.2) that capital tends to be stratified because, whilst valorisation is a continuous process, the accumulation of capital in means of production is a discrete “lumpy” process, as is entry to a new branch of production. Therefore the capital embodied in plants is dated differently. But as techniques and labour-productivity change continually over time, dated stratification is characterised according to these factors. And, as there is a tendency towards uniform prices in a market, this dated stratification is also a stratification of different rates of profit. (Thus the state of the economy conceptualised is not one of equilibrium, nor of perfect competition.)18
The prevalence of this stratification is derived from the compulsion towards valorisation, accumulation and preservation of capital. Therefore, when new techniques of production are available (with higher calculated plant rates of profit) the preservation of capital already accumulated may prevent immediate moves towards investment in new-technique and maximum rate of profit plants. Scrapping of plants is only enforced when prices no longer cover prime costs. Before that, the scrapping of plants in favour of investment in new ones is determined, firstly, by the difference in rates of profit on the investment in an already existing plant, and on that in a new plant (inclusive of the capital foregone because of scrapping); and, secondly, by the availability of means of finance (out of amortisation and/or out of additional credit). (This implies that a maximum rate of profit can only be gained by fully amortised capitals. Note that this conceptualisation here and in the remainder of this section differs from neo-classical vintage models.)19
Thus capital invested in a new plant and added to the stratification operates with up-to-date techniques of production – those with the highest VCC, maximal productivity of labour and minimal unit costs of production. Prior to the scrapping of plants, this investment increases the branch (or economy) production capacity, which induces one, or a combination of two, effects. The first is that plants in the branch operate at over-capacity as compared with the previous period; the second is that prices are driven downwards. In either case plants at the bottom of the stratification that no longer cover prime costs will have to be scrapped.20 Thus when plant (n+1) is added to the stratification (1,…,n), and when h plants are scrapped, the previous stratification (1,…,n) becomes (1+h,…,n, n+1). We shall pursue the alternative of price decreases (the over-capacity alternative has the same result in terms of the current argument; is has however different effects with respect to the employment of labour and effective demand). Because of a price decrease, the revenue of the remaining part of the previous stratification (1+h,…,n) decreases, whereas the revenue of the new stratification (1+h,…,n,n+1) typically increases with the average rate of growth.21 The decreased revenue of the capitals in the previous stratification reflects their devalorisation, which is thus due to the labour productivity of any one capital in a particular period, relative to the average, lagging behind that in the previous period.22
Therefore, because investments and costs are unaffected whilst revenue decreases, the rate of profit of the capital accumulated in the remaining part of the previous stratification (1+h,…,n) decreases. That of the capital invested in the new plant (n+1) tends, at the new price to increase as compared with the average rate of profit (1,…,n) at the previous price, or with the rate of profit of the plant just below it in the stratification, n, at the previous price.23 Since the new plant (n+1) operates at lower production costs than the previous plant (n), then in any case the rate of profit of the new plant capital at the new price is above both that of the nth and the average rate of profit. (This is in fact sufficient for the argument.) Because with the additional plant the average VCC tends to increase, the average rate of profit tends to decrease.24 But it is because of the relatively greater labour productivity of the capital added to the stratification (n+1) that its comparative profitability increases, since the value productivity of labour in the (1+h,…,n) plants thereby decreases (typically by a decrease in output prices). Therefore, not only is the money expression of labour, mi, stratified increasingly from (1,…,i,…,n), but it also tends to decrease (devalorisation) for all i when the stratification is extended.25
Thus, whilst the rate of profit on any one capital – previously invested in another branch or previously invested lower in the stratification – tends to increase, the average rate of profit in a branch or in the economy as a whole, tends to decrease. The contradiction of the tendency of the aggregate rate of profit to fall, and the increase in the rate of profit along with the accumulation of capitals in new plants, is thus resolved in the stratification of capitals.
This conceptualisation of capital stratification is, of course, not restricted to the tendency for the VCC to increase. Rather, it is the typical form of existence of capital, and therefore it constitutes also the conditions for the reproduction of the tendency of the rate of profit to fall. However, its effect on the average rate of profit will increase along with the increase of the relative share of fixed capital. To be sure, capital stratification by itself does not prove the existence of the TRPF (this would require a more encompassing study), but the account of stratification shows that the TRPF may consistently be reproduced by compatible “microeconomic behaviour”.
7 The TRPF and devaluation of capital
It was shown in the previous section how the stratification of capital provides adequate conditions for the existence of the TRPF, thus resolving its contradiction. But in this resolution the contradictions of the TRPF find further expression. One expression is the tendential negation of capital stratification (see §8). The other is that the price decrease typically associated with this form of accumulation, since it also affects input prices, may prevent the increasing technical composition of capital (TCC) being translated into an increasing value composition of capital (VCC).26
However, this price decrease affects the fixed capital outlay only of new plants, thereby decreasing the VCC when the top of the stratification TCC is replicated. Thus, as far as fixed capital outlay is concerned, this counteracting tendential price decrease does not affect the fall in the rate of profit of the prevailing capital, which is the effect of devalorisation of capital i.e., a relative decrease in valorisation of the previous stratification of capitals (§6). Input price decreases affect rather the value of the capital outlay of the previous stratification, and this devaluation of capital tends to be carried into effect by output price decreases. That is, when input prices decrease – affecting new plant investments – and when this is translated into output prices, it again decreases the revenue of the previous stratification of capital. Therefore there are two ways (depending on the accounting practice – historical or current cost accounting) in which devaluation of capital may be manifest. One is in a further decrease in the rate of profit, the other is in the immediate writing-off of capital when prices of the means of production decrease. In both cases the sum of depreciation allowances and surplus-value (i.e., the cash-flow) decreases.
From a one-sided physicalist (use-value) approach it might seem that such devaluation of capital does not affect its reproduction – in particular when (in case of current cost accounting) capital is devalued immediately so that the level of the profit rate is restored to that just prior to devaluation. Indeed physical reproduction (that is, the number of units of output of a plant) need not be affected by the input price decrease because new means of production can be bought at the lower price. But this does not take away the fact that because of devaluation the valorisation potential has decreased. This becomes obvious when a plant is wholly financed by credit: then the amortisation fund may be sufficient to buy a new plant, but not to cancel the credit.
Thus one expression of increasing labour productivity, TCC and instantaneous VCC is the devaluation of existing capital. Macroeconomically the devaluation of capital is the counterpart of the price decrease that counteracts the translation of the increase in the TCC into an increase in the VCC.27
8 The TRPF and the restructuring and centralisation of capital
The more rapid and sustained the productivity increase, the more it has the effect of wiping out the mass of profit of unamortised capitals, as expressed in both the decrease in the rate of profit along with the devalorisation and scrapping of plants (§6), and the devaluation of capital (§7). The speed of this productivity increase is contingent. Should it accelerate, then the rate of liquidation of the least efficient plants will also accelerate. Prices established in the market then approach the level of prices implied by the most efficient plants, which further devalues capital. With an accelerating rate of liquidation the forces giving rise to the TRPF tend to be expressed in the restructuring of capital, one form of which is the bankruptcy of capitals and the possible repurchase of liquidated assets by other capitals at near-to scrap value. Another form, which may prevent such bankruptcy is that of mergers, take overs and participations, that is, the centralisation of capital. Restructuring and centralisation reduce the range of the stratification of capital. Because stratification is a condition for existence of the TRPF, range reduction is a further expression of the contradiction of the TRPF.
Centralisation of capital counteracts the tendency for the TCC (and VCC) to rise; it tends to retard technical change, the implementation of which would require the building of new plants. Stratification proceeds by the temporary creation of over-capacity, and price reduction such that the least efficient plants are expelled from the stratification. This is only feasible (without a considerable reduction in profit and the rate of profit for the initiating capital) with a sufficiently large difference in productivity of labour and unit costs of production between the top and the bottom plant. With the reduction in the range of the stratification because of centralisation, it is exactly this difference in costs which is reduced. Price decreases may then not lead to scrapping of plants, so that additional plants then tend to increase over-capacity.
However, innovation in new techniques and their implementation in additional plants may still be profitable if they create a sufficient cost difference. This may require technological and technical knowledge to build up, so that technical change would then tend to come in waves. During such a build-up, the scrapping of plants will stagnate, but once sufficient technical knowledge has accumulated, the stratification will be extended again etc., ultimately giving rise to renewed devaluation and centralisation.
Thus the contingent expression of the forces generating the TRPF, in centralisation of capital, counteracts the existence of the TRPF in capital stratification. So the rise of the VCC itself is counteracted. Because this has a wave-like character the TRPF is manifest in the form of cycles of centralisation. (Note again that the term cycle is used in the sense of fluctuation in general, without any specification as to duration or regularity.)
This concept of the process of “restructuring of capital” does not, in general, diverge from the mainstream marxist tradition. Fine and Harris in particular (1979, pp. 83–7) link the process of restructuring to the TRPF. An excellent treatment in this respect is given by Weeks (1981, pp. 208–13). The notion of “centralisation of capital” on the other hand does differ from that of the mainstream, in particular in the way the wave-like expression of the TRPF is theorised. The notion that technical change comes in waves is akin to Schumpeter’s concept (see e.g. his 1943), but he offers no account as to why it is that inventions are produced during the slump and implemented at the beginning of and in the upturn. Within the framework presented here the argument is that inventions are produced throughout, but that their implementation in the slump would not pay, not just because there is a slump (since, as presumably over-capacity has been cured via restructuring, slack demand cannot be the problem), but because the centralisation of capital produces a decrease in the range of stratification, thus in fact competition between capitals remains only latent.
After a wave of centralisation of capital within a branch of production, new capitals may well enter the branch. At a high level of abstraction such flows of capital are induced by the tendency towards equalisation of average rates of profit (§2). More concretely, and in line with the presentation above, the technological and technical knowledge adopted in one branch, may profitably be applied in another – thus establishing new gaps in, and extending the range of, the stratification.
Summary and conclusions28
The TRPF contradicts the tendential increase in profit and the rate of profit associated with the accumulation of capital in new plants. This contradiction, however, has usually been conceived as an inconsistency, and so authors such as Okishio and Roemer have rejected the TRPF because of its lack of “microeconomic foundations”. It has been shown that the TRPF may be adequately grounded if the comparative static analysis of these authors is replaced by a dynamic conceptualisation, incorporating fixed capital.
The contradiction is transcended in the existence of stratification of capital. It was argued that when new VCC increasing techniques of production are available (with higher productivity of labour and lower unit costs of production), preservation of capital already accumulated prevents immediate moves towards investment in plants embodying such techniques, and so capital tends to be stratified in a range of different compositions of capital, money expressions of labour and associated rates of profit. When additional capital is accumulated in new plants, the resulting capacity increase produces either a price decrease or over-capacity. Because of the resulting revenue decrease, sub-marginal plants at the bottom of the stratification will have to be scrapped. Therefore, because of the higher productivity of new plant capital relative to the previous stratification, its comparative profitability increases, along with a comparative profitability decrease (devalorisation) of the rest of the stratification. Therefore, whilst the rate of profit of newly accumulated capital may increase relative to the capital just below it in the stratification, even if its value composition is higher, the average rate of profit of the stratification decreases because of the average increase in the value composition of capital. Thus whilst the average rate of profit decreases, profit is “redistributed” from the bottom to the top of the stratification.
The price decrease associated with TRPF was shown to counteract the VCC increase of newly accumulated capital, but at the same time the price decrease of the means of production produces a devaluation of the capital previously accumulated in fixed means of production. This reduces the reproduction of accumulation. (General over-capacity instead of price decreases produces a similar result, although the VCC increase is then not counteracted.) A further, but contingent, expression of the TRPF lies in the restructuring and centralisation of capital. The more rapid and sustained the productivity increase, the more it has the effect that plants will have to be scrapped before the capital invested has been amortised. With accelerating liquidation of the least efficient plants, the forces giving rise to the TRPF are expressed in the restructuring of capital, one form of which is the centralisation of capital. The range of the stratification of capital is thereby reduced, counteracting the gradual increase in the VCC. Technical knowledge then tends to be implemented in waves, and therefore in this expression the TRPF is manifest cyclically.
The tendency towards over-accumulation of capital (§4) and the TRPF are tendencies at the same level of abstraction. If we blend out their systematic interrelation then they would both give rise to a tendential decline in the rate of accumulation and generate counteracting forces. Both also reveal, in their different ways, the contradictory nature of capitalism. The forces of valorisation and accumulation of capital are expressed in forms which appear to contradict their existence. Valorisation of capital tendentially gives rise to over-accumulation and therefore decrease of valorisation (expressed in labour-shortage profit squeeze but also underconsumption). Similarly the drive to increase the profit rate of the individual capitals tendentially gives rise to a decrease in the rate of profit of capital as a whole (expressed in devaluation, restructuring and centralisation of capital).
What are the conclusions that can be drawn?
(1) Adequate “microeconomic foundations” for the TRPF may be provided, independently of a change in the rate of surplus-value (§6). The generality of the Okishian theorem has thus been refuted, and that theorem is thus confined to a special case (i.e., neoclassical-perfect competition in the absence of fixed capital). The tendential increase in the VCC and the derived TRPF thus stands as an important economic force in explaining concrete economic development.
(2) Does this imply that the rate of profit must fall (i.e., must there by implication be an empirically observable trend fall in the rate of profit)? No, for the following reasons (3)–(6).
(3) The devaluation of capital associated with the TRPF is concretely (and empirically) expressed either in the writing off of capital (affecting the VCC) or in a fall of the rate of profit; which one of the two depends on the contingent accounting practice (§7).
(4) A further concrete expression of the TRPF is in the restructuring and centralisation of capital, whence the TRPF is manifest cyclically (§8).
(5) The TRPF interacts with the tendency towards over-accumulation of capital; one expression of this tendency is a (cyclical) change of the rate of surplus-value, as derived from inter alia the reserve of labour (§4).
(6) The issue is complicated even more due to the interconnectedness of all this with the credit system and e.g. processes of inflation (and the latter cannot be simply levelled out by indexing). But further, this amalgam is modified by the outcome of economic policy. (See Reuten and Williams 1989, respectively chs. 5 and 9.)
I do not agree with Roemer that “empirical investigations … cannot provide refutation of a theory” (see §5). Simple theories could simply be refuted. But I also do not agree with Reati (1986, p. 56) when he says that because “theoretical debate [concerning the TRPF] has not really settled the matter … only empirical tests can lead to further progress”. Empirical investigation is surely in need. But it makes no sense to test a theory that has not been fully constructed. And a full construction would need to encompass points (3)–(6) above.
I would like to thank Michael Williams and Jörg Glombowski for their helpful comments on earlier versions of this paper and Marilyn Gruschka for correcting my English. The paper has also benefitted from comments made by two anonymous referees and the editors of the CJE. Any remaining errors are, of course, mine. A more extended discussion of the issues raised in this article is in Reuten and Williams (1989), as specified at the appropriate places.
See, however, the references in Harris (1983, p. 311) and Duménil, Glick and Rangel (1987a, pp. 332–3).
Recent empirical investigations of the rate of profit and related variables include Duménil, Glick and Rangel (1987a, 1987b), Lipietz (1986), Reati (1986), Webber and Rigby (1986), and Wolff (1986). The upshot of the present article is, first, that the tendency at hand not only interacts with other tendencies of accumulation (§4), but, secondly, that the expressions of this tendency are likely to be manifest in a cyclical manner (§7–§8).
Thus the concept of value does not refer to labour-time (as it would in a ‘labour-embodied’ theory of value). Nevertheless, because the dimension of some unit of labour is time, the value of labour (a concept incompatible with a Marxist labour-embodied theory) refers to some labour-time. Its quantitative expression is always in terms of a standard of money (e.g. £x or $y). Thus, the value of (one hour of) labour (time) in some plant is, e.g., £12 or $18 (indicated below by the symbol m). The concept ‘labour-power’ is the conventional Marxist one, i.e., the ability to perform labour for a definite period of time. Again the quantitative expression of the ‘value of labour-power’ per unit of time is in terms of a standard of money, e.g., £7 (i.e., the wage rate).
This concept ‘money expression of labour’ (m´ and m´i), differs from Aglietta’s “monetary expression of the working hour” (1976, p. 43; see also De Brunhoff 1976, p. 39) which we denote by m*. The latter is used by Aglietta only at the macroeconomic level and it is predicated upon what he calls “the monetary constraint” that realisation is equal to production. Thus Aglietta’s m* = Y/l is the monetary condition that Y = Y´. My concept ‘money expression of labour’ is the reciprocal of Foley’s (1986, p. 15) “monetary expression of labour”.
The reserve of labour-power is determined by accumulation, demographic and social-demographic factors. Formally one can, of course, not exclude the possibility of a rate of surplus-value such that the rate of capital accumulation will be reduced to the growth rate of labour supply; for a Marxist account of this state see Glombowski (1983, pp. 378–81).
The other expression of over-accumulation of capital is in underconsumption (see Reuten and Williams 1989, ch. 3).
The theorisation of these or similar effects of labour shortage is part and parcel of almost every strand in economic theory, though there are important theoretical differences concerning the reactions to wage increases. Within the Marxist tradition (not excluding more Ricardian oriented authors) this theory has been developed at various levels of abstraction and from various methodological perspectives into what is labelled the ‘theory of the profit squeeze’. In general, the focus in this theory is on class struggle over the distribution of income and its effect on accumulation and employment. The (initially) more empirically-oriented work derives from Glyn and Sutcliffe (1971, 1972) and Boddy and Crotty (1975). Model-analytic work derives in particular from Goodwin (1967) (for a short taxonomy and additional references, see Glombowski 1984, pp. 74–5; see also Goldstein 1985).
As for the conceptual differences between the TCC, OCC (organic composition of capital) and VCC see Orzech and Groll (1989) and Reuten and Williams (1989, pp. 119–21). I refrain from using the term OCC because it is often associated with a ‘labour-embodied’ concept of value.
Thus whilst these tendencies interact, the one tendency is not merely a reaction to the other. Both these tendencies are continually operating forces. See also Hunt (1983, pp. 137–9).
However, such price decreases do affect the composition of capital through its means of production component (see section 7).
Thus in the absence of inflation the effect of the increased number of use-values will tend to be offset by a decrease in the general price level (cf. Aglietta 1976, ch. 6).
See Reuten and Williams (1989, ch. 5), for the articulation of the TRPF and changes in the rate of surplus-value, as well as with the tendency to over-accumulation in general. See also Lipietz (1986, pp. 15–16).
Various authors have generalised and/or tried to improve on the Okishio theorem (besides Roemer, e.g., Alberro and Persky 1979; Van Parijs 1980, 1983; Bowles 1981; Harris 1983; Coram 1986; and Bidard 1988), resulting in a more or less qualified rejection of the TRPF. Others, on the other hand, have shown that a combination of relaxation and restriction of the Okishian assumptions (in particular when the rate of surplus-value is held constant, or when fixed capital is introduced, or when competition is not neoclassical-perfect) leaves open the possibility of a fall in the rate of profit when the TCC increases (see Shaikh 1978, 1980; Salvadori 1981; Weeks 1981 (ch. 8), 1982; Fine 1982 (ch. 8); Laibman 1982; Hunt 1983; Negishi 1985 (ch. 4); Foley 1986 (ch. 8); Lipietz 1986; Schutz 1987; Moseley 1991, ch. 1–2). In general, the latter contributions have either reduced the TRPF to a special case, or left the ‘microeconomic foundations’ somewhat in the open.
When the constant real wage is conceived of as a commodity input (Roemer 1979, pp. 381–2) then the rate of profit always increases if the input-output matrix is indecomposable; if it is decomposable the rate of profit might remain the same.
Shaikh’s (1978) critique of Okishio and Himmelweit – to which Roemer (1979) also replied – is that capitalists are forced (microeconomically) to reduce cost-prices whereby, indeed, according to the Okishio theorem, the profit margin on costs increases; at the same time, however, the profit rate might decrease. This is unconvincing, because it seems to imply that some fixed capital costs are not part of cost-prices. What is more, good arguments can be provided for the thesis that the profit rate is the first criterion from which the other two follow. See also the comments on Shaikh by Steedman (1980), Nakatani (1980), Bleaney (1980), Fine (1982, pp. 125–7) and Van Parijs (1983), as well as the reply to the first three by Shaikh (1980).
This is all the more remarkable because Roemer (rightly) accuses Shaikh (1978) of neglecting amortisation of fixed capital. It may be noted that this problem would remain even if technical changes could be perfectly foreseen (pp. 387–8), unless, of course, capitalists were, because of foreseen technical changes, to decide not to invest at all.
In the case of perfect competition, capital would always immediately move to the new-technique plant (Salter 1960). An earlier presentation of the stratification framework set out here is in Reuten (1978, pp. 19–26). For a similar conceptualisation in this respect see Weeks (1981, pp. 204–8).
In the neo-classical concept, obsolescence of plants is determined by the real wage (wage costs exceeding the average labour-productivity on a plant), rather than by the addition of plants to the stratification, introducing new cost-reducing techniques of production and by (as explained below) the resulting price decrease and/or over-capacity.
The new plant capital may initiate this price decrease because it operates at minimal costs. It then functions as price leader. Subsequent scrapping is the typical process. Of course, plants at the bottom of the stratification may be scrapped pre-emptively. The expected price decrease, or alternatively the expected over-capacity will be taken into account by the capital considering the new plant prior to the investment. In general, the initiation of a price decrease by this capital, in order to induce scrapping of plants at the bottom of the stratification so that the new plant can operate at near to full capacity, may be advantageous to it. Although I think that it is important to conceptualise the processes of adaptation (price decrease and scrapping) the argument here does not rely on it. In principle the argument (price decrease in particular) could also be cast in terms of a Sraffian equilibrium model with joint production. Whilst the reference here is to a branch, the argument typically holds for the economy as a whole. A formal but fairly simple model of the argument is presented in Reuten and Williams (1989, appendix to ch. 4).
Of course in the case of (macroeconomic or branch) stagnation revenue may remain constant or decrease. On the other hand, one branch may, of course, grow above average. This does not, however, affect the general argument.
Devalorisation of capital goes beyond any normal wear and tear. (Note the difference between the devalorisation of capital as introduced here, and the devaluation of capital as introduced in the next section.) It might be argued that to the extent devalorisation is foreseen at the point of investment, it is incorporated in calculating the ‘marginal efficiency of capital’. But even if there were perfect foresight in this, the argument is unaffected. It cannot prevent devalorisation, and even with devalorisation net profits over the lifetime of the asset may still be positive and optimal.
Within the ‘Okishian’ argument it is sufficient that the (n+1) composition of capital and rate of profit at the old prices is just above that of (n) at the old prices. So if the (n+1) rate of profit is only just above that of (n), and if the (1+h,…,n) rate of profit decreases at the new prices or capacity utilisation, then the Okishian argument is falsified for all plausible cases (those in which there is indeed accumulation of capital).
Price decreases, of course, affect capital outlay on circulating constant capital (raw materials etc.). It does not, however, affect previous capital outlay on fixed constant capital. The effect of the related devaluation of fixed capital is taken up in the next section.
The notion of the extra profits gained by the capitals at the top of the stratification is closely related to Schumpeter’s notion of temporary monopoly profits accruing to the first capital to innovate, which are gradually eroded as the innovation diffuses through the industry and even the economy (see, e.g., Schumpeter 1943).
This effect depends upon the argument that price changes do not directly affect the money wage rate, and that the latter is determined rather in relation to the rate of unemployment. Therefore input-price decrease affects the value composition of capital (even if the price decrease were to be similar in all branches of the economy) rather than the rate of profit.
It could be argued that within the conceptualisation of capital stratification presented in section 6, devaluation of capital also occurs, though only latently, when new techniques are introduced (prior to any input price decrease). Then, depending on the accounting practice, devalorisation may be manifest either in a decreasing money expression of labour (mi), or in a reduction of the value of capital. In the first case valorisation decreases in comparison with the anticipated valorisation; in the second case previous valorisation is partly annihilated. As the cash-flow is not affected the net effect is, of course, the same. For expositional purposes I have referred only to the first manifestation.
[2023 note] The following abbreviations are used. TRPF: theory of the tendency of the rate of profit to fall (Introduction and §4.2); VCC: value composition of capital (§4.2); TCC: technical composition of capital (§4.2).
References
Aglietta, Michel 1976, Régulation et crises du capitalisme, Calmann-Lévi, English translation by D. Fernbach, A theory of capitalist regulation: the US experience, London: NLB, 1979.
Alberro, Jose, and Joseph Persky 1979, ‘The simple analytics of falling profit rates, Okishio’s theorem and fixed capital’, Review of Radical Political Economics 11, Fall.
Bidard, Christian 1988, ‘The falling rate of profit and joint production’, Cambridge Journal of Economics 12(4): 355–60.
Bleaney, Michael 1980, ‘Maurice Dobb’s theory of crisis: a comment’, Cambridge Journal of Economics 4(1): 71–3.
Boddy, Radford, and James Crotty 1975, ‘Class conflict and macro-policy: the political business cycle’, Review of Radical Political Economics 7, Spring: 1–19.
Bowles, Samuel 1981, ‘Technical change and the profit rate: a simple proof of the Okishio theorem’, Cambridge Journal of Economics 5(2): 183–6.
de Brunhoff, Suzanne 1976, État et capital, Presses Universitaires de Grenoble & Maspero; English translation by M. Sonenscher, The State, Capital and Economic Policy, London: Pluto Press, 1978.
Coram, Bruce T. 1986, ‘Marx, Roemer and the theory of the falling rate of profit’, Australian Economic Papers 25, December: 265–71.
Cutler, Antony, Barry Hindess, Paul Hirst, and Athar Hussain 1977, Marx’s Capital and capitalism today, Vol. I, London: Routledge & Kegan Paul.
Duménil, Gerard, Mark Glick, and Jose Rangel 1987a, ‘The rate of profit in the United States’, Cambridge Journal of Economics 11: 331–59.
Duménil, Gerard, Mark Glick, and Jose Rangel 1987b, ‘Theories of the Great Depression: why did profitability matter?’, Review of Radical Political Economics 19(2): 16–42.
Fine, Ben 1982, Theories of the capitalist economy, London: Edward Arnold.
Fine, Ben, and Laurence Harris 1979, Rereading Capital, London: Macmillan.
Foley, Duncan 1986, Understanding capital: Marx’s economic theory, Cambridge, MA: Harvard University Press.
Glombowski, Jörg 1983, ‘A Marxian model of long run capitalist development’, Zeitschrift für Nationalökonomie/Journal of Economics 43(4): 363–82.
Glombowski, Jörg 1984, ‘Kritische Kommentare zur Akkumulationstheorie’, Mehrwert 25: 67–80.
Glyn, Andrew, and Bob Sutcliffe 1971, ‘The critical condition of British capital’, New Left Review 66.
Glyn, Andrew, and Bob Sutcliffe 1972, British capitalism, workers and the profit squeeze, Harmondsworth: Penguin Books.
Goldstein, Jonathan P. 1985, ‘The cyclical profit squeeze: a Marxian microfoundation’, Review of Radical Political Economics 17(1 and 2): 103–28.
Goodwin, Richard M. 1967, ‘A growth cycle’, in Capitalism and economic growth, edited by C.H. Feinstein, Cambridge: Cambridge University Press, pp. 54–8; revised and enlarged in E.K. Hunt and J.G. Schwartz (eds), A critique of economic theory, Harmondsworth: Penguin Books, 1972, pp. 442–9.
Harris, Donald J. 1983, ‘Accumulation of capital and the rate of profit in Marxian theory’, Cambridge Journal of Economics 7(4): 311–30.
Himmelweit, Susan 1974, ‘The continuing saga of the falling rate of profit: a reply to Mario Cogoy’, Bulletin of the Conference of Socialist Economists 9: 1–6.
Hunt, Ian 1983, ‘An obituary or a new life for the tendency of the rate of profit to fall?’, Review of Radical Political Economics 15(1): 131–48.
Jevons, W. Stanley 1871, The theory of political economy, reprinted 1970, Harmondsworth: Penguin Books.
Laibman, David 1982, ‘Technical change, the real wage and the rate of exploitation: the falling rate of profit reconsidered’, Review of Radical Political Economics 14(2): 95–105.
Lipietz, Alain 1986, ‘Behind the crisis: the exhaustion of a regime of accumulation: a “Regulation School” perspective on some French empirical works’, Review of Radical Political Economics 18(1 and 2): 13–32.
Moseley, Fred 1985, ‘The rate of surplus-value in the post-war US economy: a critique of Weisskopf’s estimates’, Cambridge Journal of Economics 9(2).
Moseley, Fred 1991, The falling rate of profit in the post-war US economy, London: Macmillan.
Nakatani, Takeshi 1980, ‘The law of falling rate of profit and the competitive battle: comment on Shaikh’, Cambridge Journal of Economics 4(1): 65–68.
Negishi, Takashi 1985, Economic theories in a non-Walrasian tradition, Cambridge: Cambridge University Press.
Okishio, Nobio 1961, ‘Technical changes and the rate of profit’, Kobe University Economic Review 7: 85–99.
Orzech, Ze’ev B., and Shalom Groll 1989, ‘Stages in the development of a Marxian concept: the composition of capital’, History of Political Economy 21(1): 57–76.
van Parijs, Philippe 1980, ‘The falling rate of profit theory of crisis: a rational reconstruction by way of obituary’, Review of Radical Political Economics 12(1): 1–16.
van Parijs, Philippe 1983, ‘Why Marxist economics needs microfoundations: postscript to an obituary’, Review of Radical Political Economics 15(2): 111–24.
Reati, Angelo 1986, ‘The rate of profit and the organic composition of capital in West German industry from 1960 to 1981’, Review of Radical Political Economics 18(1 and 2): 56–86.
Reuten, Geert 1978, ‘A concretization of the operation of the general rate of profit (on the problem of the alleged transformation of values into prices of production)’, Research Memorandum no. 7801, Faculty of Economics, University of Amsterdam.
Reuten, Geert 1988a, ‘Value as social form’, in Value, social form and the state, edited by Michael Williams, London: Macmillan, pp. 42–61.
Reuten, Geert 1988b, ‘The money expression of value and the credit system: a value-form theoretic outline’, Capital & Class 35: 121–41.
Reuten, Geert, and Michael Williams 1989, Value-form and the state; the tendencies of accumulation and the determination of economic policy in capitalist society, London: Routledge.
Roemer, John E. 1979, ‘Continuing controversy on the falling rate of profit: fixed capital and other issues’, Cambridge Journal of Economics 3(4): 379–98.
Salter, Wilfred E.G. 1960, Productivity and technical change, Cambridge: Cambridge University Press.
Salvadori, Neri 1981, ‘Falling rate of profit with a constant real wage: an example’, Cambridge Journal of Economics 5(1): 59–66.
Schutz, Eric 1987, ‘Non-produced inputs, differential profit rates and the Okishio theorem’, Review of Radical Political Economics 19(2): 43–60.
Schumpeter, Joseph A. 1966 [1943], Capitalism, socialism and democracy, London: Unwin University Books.
Shaikh, Anwar 1978, ‘Political economy and capitalism: notes on Dobb’s theory of crisis’, Cambridge Journal of Economics 2(3): 233–51.
Shaikh, Anwar 1980, ‘Marxian competition versus perfect competition: further comments on the so-called choice of technique’, Cambridge Journal of Economics 4(1): 75–83.
Shaikh, Anwar 1987, ‘The falling rate of profit and the economic crisis in the US’, in The imperiled economy, book I: macroeconomics from a left perspective, edited by Robert Cherry et al., New York: URPE.
Steedman, Ian 1980, ‘A note on the “choice of technique” under capitalism’, Cambridge Journal of Economics 4(1): 61–4.
Webber, Michael J., and David L. Rigby 1986, ‘The rate of profit in Canadian manufacturing, 1950–1981’, Review of Radical Political Economics 18(1 and 2): 33–55.
Weeks, John 1981, Capital and exploitation, London: Edward Arnold.
Weeks, John 1982, ‘Equilibrium, uneven development and the tendency of the rate of profit to fall’, Capital & Class 16: 62–77.
Wolff, Edward N. 1986, ‘The productivity slowdown and the fall in the US rate of profit, 1947–76’, Review of Radical Political Economics 18(1 and 2): 87–109.