David Yaffe - The Marxian Theory of Crisis, Capital and the State
(a) Capital production
What distinguishes Marx from his Classical predecessors is that he never loses sight of the fact that the 'value-producing' process, central to capitalist production, is only an historical form of the material production and reproduction process of society. The labour process becomes a 'value-producing' process and the social relations are transformed into economic categories under capital production. Capitalist production is oriented not towards consumption needs, this is the production of use-values, but towards production for profit, that is the production of exchange-values. It is the dual nature of a commodity under capitalist production conditions, that is as a use-value and exchange-value, that constitutes the most general contradiction of the capitalist system. This may be put in another form: while the labour process is only limited by the natural resources available, by the historical stage of development of the social productivity of labour and the mass of labour in society, the labour process as a 'value-producing' process has much narrower limits. Under capitalist production natural resources are only utilised, the social productivity of labour only developed, labour is only employed if it serves the self-expansion of capital, i.e. the reproduction of the existing capital values and the creation of additional value, surplus-value. Capitalist production therefore, is the production of exchange-values through the production of commodities, the objective being surplus-value as additional exchange-value.
Accumulation is the continuous process of reproduction and self-expansion of capital (Verwertungsprozess des Kapitals), it is the reproduction of capital on a progressively increasing scale. Necessarily, accumulation is also the reproduction of the capitalist social relation on a progressive scale, 'more capitalists or larger capitalists at this pole, more wage workers at that' So long as capitalist relations of production exist, so long as one class owns the means of production as capital, and another has to sell its labour-power to live, so long will the end and aim of production be the accumulation of capital.
Before an analysis of the accumulation process itself is possible it is necessary to say something about Marx's methodology and, in particular, about the concepts 'capital in general' (Kapital im allgemeinen) and 'many capitals' (vielen Kapitalien) or capital in its 'real' form of competition.
(b) 'Capital in general' and 'many capitals'
It is the particular form that social relations take under capitalist production, their fetishistic form, which makes it all the more necessary for political economy to start from simple (abstract) conceptions and move by a process of increasing concretisation to grasp the concrete reality. 'The method of advancing from the abstract to the concrete is but the way of thinking by which the concrete is grasped and reproduced in our mind as concrete'  The failing of what Marx called 'vulgar economy' is that it remains in the 'estranged outward appearances of economic relations'. In so doing it takes on an apologetic character for the given social relations and by treating them as 'eternal' fails to grasp their contradictory character. 'But all science would be superfluous if the outward appearance and essence of things directly coincided', and it is precisely this point that indicates the indispensability of the value analysis for Marx's total work. .
Marx begins this analysis in Volume I of Capital by examining the exchange of commodities. From this he deduces that Value or expenditure of abstract human labour lies at the basis of commodity-exchange, independent of the use-values of the commodities. He then moves on to the only form in which the value of commodities can be expressed, that is exchange-value. After deducing the money form of value he goes on to analyse the capital form of value. It was the examination of capital, of value that generates surplus-value (value in process ), which presupposes a definite historical relationship, the wage-labour relationship (labour power as a commodity), that was to take up the major proportion of Marx's effort and work.
In order to develop the concept of capital it was first of all necessary to abstract from 'many capitals' or the action of capitals on one another through competition. The latter would be analysed after the consideration of what they (many capitals) have in common, as capital.  In the Grundrisse the point is made very clear:
'Capital, in so far as we are considering it here ... is capital in general. . Value, money, accumulation etc. prices etc. are presupposed, just as is labour, etc. However we are concerned neither with a special form of capital nor with individual capital as differentiated from other individual capitals, etc. We are remaining with its process of generation and growth (Entstehungsprozess). This dialectical process of generation and growth is nothing but the ideal expression of the real movement of capital. The later relationships are to be regarded as a development from this 'central core' (Keim). ...'. 
It follows that the later form of capital is contained in embryonic form (Keimform) within the general concept of capital. This means not only the 'civilising' dynamic tendencies of capital but also the latent contradictions which drive capital beyond its own -limits. 
Marx refers to 'capital in general' sometimes as 'the capital of the whole society' or nation, or as the general economic basis of one class as opposed to another class. 'Capital in general' -is not a mere abstraction or an arbitrary abstraction, but an abstraction that must be understood as the differentia specifica of capital in distinction to all other forms of wealth. If we are to comprehend the basic presupposition of the capital relation-the relation of capital and labour and the role of surplus-value as the driving force of capitalist production-then we must begin our analysis with 'capital in general' undisturbed by a consideration of 'many capitals' or the actions of capitals on one another.
A scientific analysis of competition is not possible before we have a conception of 'capital in general', that is, of the 'inner nature of capital'. Competition is the 'essential locomotive of bourgeois economy' which nevertheless does not create or establish its laws but merely allows them to be realised ('allows them to be exhibited but does not produce them'). Capitalist production exists in its most 'adequate' form in so far as free competition develops. Nevertheless, as soon as capital feels itself threatened it will 'seek refuge in other forms', which appear to perfect its rule as capital 'through curbs on free competition'. Marx offers here a clear context for understanding capitalism in its 'latest' stage. It is precisely the nature of capital itself and the preservation of its rule as capital that is central for an understanding of capitalism in its 'monopoly' stage. It has not ceased to be capital by virtue of the 'curbs on free competition'. On the contrary it is precisely the 'rule of capital' that makes the 'curbs on free competition' necessary. It follows, therefore, that the analysis of 'capital in general' is still the starting point of any analysis of contemporary capitalism. The value analysis of Volume I of Capital still retains its validity, and although the 'law of value' is 'modified', it nevertheless is the basis for any serious consideration of modern capitalism. The 'modifications' and this includes interventions by the capitalist State, are within limits governed by the preservation of the 'rule of capital' itself and so modern capitalism, like the capitalism of Marx's day, is subject to value-analysis.
Baran and Sweezy are, therefore, wrong when they assert:
'the Marxian analysis of capitalism still rests in the final analysis on the assumption of a competitive economy'.
It rests on an analysis of 'capital in general', undisturbed by considerations of competition, that has as its basis the value relations examined in Volume I of Capital. As long as capital and therefore capitalist production relations exist whether the market structure is competitive or monopolistic, the value analysis is central.
Volume I and II of Capital are primarily concerned with an examination of 'capital in general' and the special forms of the existence of 'capital in general' as fixed and circulating capital. That this is often not understood has lead to all kinds of confusion when considering Marx's theory of crisis and its relevance for an examination of contemporary society. It is not surprising that so many critics of Marx have talked of a contradiction between Volume I and III of Capital (e.g. Bohm-Bawerk); and others have confused 'capital in general' with capital in 'reality', 'many capitals' (e.g. Rosa Luxemburg). It is only in Volume III of Capital that Marx begins to 'locate and describe the concrete forms which grow out of the 'movements of capital as a whole' as they approach step by step the form which they assume on the surface of society, in the action of different. capitals upon one another, in competition, and in the ordinary consciousness of the agents of production themselves'. 
(c) Productive and unproductive labour
This distinction, so central to Marx's theory of accumulation, is one often misunderstood by both Marxists and Marx's critics. No discussion of the role of State intervention is possible until this distinction is clarified.
A productive labourer is one who works for the self-expansion of capital and produces surplus-value for the capitalist through the production of commodities . It is labour which is directly exchanged with capital for the purpose of augmenting capital . The use-value of the commodity, in which the labour of a productive labourer is embodied, is in no way relevant to this definition; the commodity 'may be of the most futile kind' . The definition of productive labour is the expression of a definite social relation of production, that is, of capitalist production. Therefore, any 'moralistic' definition of productive labour, i.e. useful to society, has nothing in common with Marx's definition and merely confuses the issue by abstracting from the particular mode of production and type of society. The goal of the capitalist production process 'is the accumulation of wealth, the self expansion of value, its increase; that is to say the maintenance of the old-value and the creation of surplus-value. And it achieves this specific product of the! capitalist production process only in exchange with labour, which for that reason is called productive labour' . .
Unproductive labour is labour that is not exchanged with capital but directly with revenue, that is with wages or profits . This means that services paid out of the wage of workers or the income of capitalists are unproductive from the capitalist standpoint. They constitute a reduction of the surplus-value that is available for reinvestment as capital. The case of services, e.g. education, which actually maintain or raise the value of labour-power or alter the skill (complexity of labour) of the labourer is slightly more complicated. Marx talks of such services as yielding in return 'a vendible commodity etc. namely labour-power itself, into whose costs of production these services enter' . But the utility of the service alters nothing as far as the economic relation is concerned, as revenue is exchanged against labour. The result of the service, by its very nature, cannot even be guaranteed by those rendering the service. . The case of State expenditure on education exactly fits this category. In so far as education increases the total value of labour-power (of all labourers) and has little or no effect on the value-forming potential of the labourer (skill etc.) besides leading to a fall in the average rate of profit, it alters the redistribution of the social product in favour of the labourers and at the expense of surplus-value. To the extent that it increases the skill etc. of the worker and/or productivity in the wage goods industries the latter effects are compensated .
Mere circulation costs from the standpoint of capitalist production are unproductive. Although wage-labour is performed and the capitalist ingesting in this sphere receives a profit, no addition to surplus-value, to total social capital, is made. Such costs, necessary for the realisation of profits reduce the overall rate of profit. The employment of commercial workers, office staff etc. increase the expenses of the industrial capitalist and therefore the mass of capital to be advanced without directly increasing surplus value. If the extra costs are Δc, then the rate of profit will be reduced from s/c to s/c + Δc.  The labourer who works in the commercial sphere still performs unpaid labour and his cost to the capitalist is the value of his labour-power:
'He creates no direct surplus value, but adds to the capitalists' income by helping to reduce the cost of realising surplus-value, inasmuch as he performs partly, unpaid labour' .
These purely commercial costs are to be separated from the costs of transport shipping, storage etc. These are regarded as part of industrial production. A change in the object of labour, in the commodity takes place. In the case of transport 'its spatial existence is altered, and along with this goes a change in its use-value, since the location of this use-value is changed' . Such labour is regarded as productive labour.
It follows from all this that from the standpoint of capitalism as a whole 'variable capital' represents only the wages of productive labourers not of the total labour force. Further, surplus-value is not equal to the total surplus-product but to the surplus product of productive labourers. So that the share of wages and profits in national income does not tell us very much about the rate of exploitation. It is quite reasonable to assume that, if the unproductive (including the non-profitable state) sector is growing at a faster rate than the productive sector, total wages as a share of national income can grow and the rate of exploitation can still rise. This is because a part of net income imputed to wages is in reality a part of surplus-value produced by productive labourers. This latter point indicates one of the problems in trying to find statistical counterparts for the Marxian categories.
To conclude this section we have the following important results. Variable capital represents only the wages of productive workers. Surplus-value is the total profit of the productive sector. Constant capital is the part of the means of production employed in the productive sector. The rate of exploitation and organic composition of capital relate to the variables as defined' above. The unproductive part of total production becomes relevant when we discuss the rate of profit. Here it will be included as a extra-cost. More capital is advanced in order to finance this unproductive sector and so the rate of profit will be correspondingly lower.
Having clarified the meaning and role of some of the central categories of Marxist political economy we now turn to the general theory of accumulation and crisis.
 See Capital Volume I (Moscow Ed., 1961) p.581. In his critique of Ricardo's Theory of Accumulation Marx continually refers to accumulation as reproduction on a larger scale, increase of capital, growth of constant capital, reproduction on an extended basis and formation of additional value. He even indicates how the mere reproduction of capital invested in the machine-building industry requires continuous accumulation in other spheres of production. See Theories of Surplus-Value Part. II (Moscow Ed) p.487, p473, p522, p485 and p480-1.
 Capital Volume I, pp. 613-4 also p. 578.
 K Marx Introduction to the Critique of Political Economy as an appendix to A Contribution to the Critique of Political Economy, translated by N.I. Stone. Chicago, Charles H. Kerr & Company, 1904, p. 292 ff.
 Ibid, p. 293-4.
 Capital Volume III (Moscow Ed, 1962) p. 797.
 Capital Volume I, p. 154.
 Grundrisse der Kritik der Politischen Oekonomie. Dietz Verlag. Berlin 1953, p. 416. A great deal of this analysis follows the arguments of Roman Rosdolsky Zur Entstehungsgeschichte des Marxschen 'Kapital' Band I. Europaische Verlagsanstalt, Frankfurt, 1969, p. 61 ff.
 Grundrisse, op. cit., p. 217, (my translation).
 Ibid, pp. 317 and 237. Also Roman Rosdolsky, 1969, p. 70.
 Grundrisse, op. cit., p. 252
 Ibid, p.735.
 Ibid, p.353.
 Capital Volume I, p. 316.
 Grundrisse op. cit. p.450.
 Ibid, pp. 544-5. Marx continues 'although (the curbs on free competition) appear to complete the mastery of capital, (they) are at the same time, by curbing free competition the heralds of its dissolution, and of the dissolution of the means of production which are based on it'.
 Monopoly Capital, Monthly Review Press. N.Y. and London, 1966, p. 4.
 Paul Mattick 'Marxism and "Monopoly Capital"' in Progressive Labour Volume 6 No. 1, July-August 1967.
 Capital Volume III, p. 25.
 Capital Volume I p509.
 Theories of Surplus-Value Part I and Grundrisse p212-3
 Ibid p 154.
 Ibid p 387-8
 Ibid p153
 Ibid p163
 Ibid p393. If the schoolmaster worked for a capitalist proprietor and education was a necessary ingredient in the reproduction of labour-power; to that extent, the schoolmaster could be regarded as productive. This presumably covers the case given by Marx in Capital Volume I, p. 509. Sometimes Marx seems to imply that labour is productive if it produces surplus-value for the capitalist only and does not add to total social capital (Theories of Surplus-Value, Part I, p. 393). This might have been what he had in mind in his productive schoolmaster example if we refer to his remarks (quoting Smith) about how little `education' enters into the cost of production of the mass of working men (Theories of Surplus-Value, Part I, p. 163). We reject such a definition and hold clearly to that more consistent with Marx's general analysis that labour is only productive so long as it augments capital (Grundrisse, pp. 212-3).
 See article by Elmar Altvater and Freerk Huisken in Sozialistische Politik Nr 8 September 1970. especially pages 82-91. The whole issue of the journal is given over to a discussion of productive and unproductive labour.
 Capital Volume III p293-4.
 ibid p.294.
 Theories of Surplus-Value Part I, p.399. See also Capital Volume III p.283.
 This is one weakness in Andrew Glyn and Bob Sutcliffe's book British Capitalism, Workers and the Profits Squeeze. Penguin books, 1972. On page 15, they say, 'The share of profits is the total amount of profits expressed as a proportion of the national income.' In another place (p. 32) they speak of the share of wages as 'total wages in industry expressed as a proportion of the value of industrial output'. But they continue, 'Sometimes the share of wages in the national income is referred to; it is the total of all wages expressed as a proportion of national income. It is the counterpart of the share of profits'. They seem to show little awareness of the central problem of productive and unproductive labour in choosing their data.
 For a similar formulation of this result see Rudi Schmiede Zentrale Probleme der Marxschen Akkumulations - und Krisentheorie Soziologische Diplomarbeit. Frankfurt/Main 1972, p. 48. It should be remarked that to clearly separate productive from unproductive labour empirically is not really a possibility. The value of the conceptual separation will be clear in the later analysis.