One of the key theses of Marx’s Capital is the primacy of production over circulation. It determines the plan of Capital: production in Volume I (the only one completed and published in the author’s lifetime), circulation in Volume II, and the unity of production and circulation in Volume III. Analytically, this hierarchy leads to the assumption of a centrality of industrial capital over commercial and financial capitals, which do not directly control production. Epistemologically, it is the basis to the claim of the superiority of Marx’s own “scientific” political economy over the “vulgar political economy” considered to be confined to the sphere of circulation.
To this day, although some Marxists insist on the relative importance of circulation, the primacy of production remains a distinctive character of Marxist analysis. In this post I will argue that the continued emphasis on production is unwarranted. Indeed, a strict analytic division between production and circulation is an error. If Marx’s insights on the way exploitation can pervade formally equal relations are to be preserved, we must substitute the primary of production with the concept of coordination.
1. Production and Circulation
It is easy for a Marxist to get so accustomed to the idea of a primacy of production as to consider it selfexplanatory. But in Capital it is not a premise but a conclusion. The distinction between production and circulation appears at the end of Volume 1, Section 2. The goal of that section is to understand how capital, i.e. value accumulating itself, is possible in a market economy where exchanged commodities are supposed to be of the same value. The well known answer is that labor power is the one and only commodity which can systematically provide its buyer more value than it costs him. This argument is the precondition for Marx’s injunction that “we take leave for a time of this noisy sphere [of circulation], where everything takes place on the surface and in view of all men, and follow them both into the hidden abode of production.” So the distinction between production and circulation relies on the distinction between two types of relations, usually structured by contract: wage relations, defined as buying the control over someone else’s labor power for some time in exchange for money, and commercial relations, defined as the exchange of any other commodity.
Analytically, Marx’s argument requires the theory of value as abstract labor time, the defects of which have long been well known. But it is also interesting to examine its empirical defects.
Since the beginning of capitalism, the transfer of surplus value towards capital has happened at least as much through commercial and credit relations as through wage relations. Commercial capital was crucial in the global North, in the phase of proto-industrialization when urban capitalists bought the finished product from a network of worker-peasants in the countryside, and after that when big textile factories subcontracted orders to small ones in what came to be known as the sweatshop system. It was perhaps even more important for workers in Asia and Africa: when they were not subject to one or another form of forced labor, they were often only connected to western capital by a multilayered network of brokers. It is true that commercial capital was temporarily overshadowed by industrial capital in the century following the publication of Capital, but commercial capital did not disappear and the importance for accumulation of subcontracting inside the global North and of dominant positions in the networks of the global value chains has become increasingly important in the neoliberal era. We find paradigmatic figures of this new commercial capital both in the old textile industry, with Nike, the first brand which never had factories, and at the other end of the spectrum, in the digital platforms that dominate social spaces by coordinating many formally independent agents. This has led some to argue that we are entering a new phase of merchant capitalism.
The implication is that exploitation happens not only through the direct buying of labor power, but simultaneously at many different embedded scales. Let us consider a very common situation where a small textile shop, probably somewhere in Asia but also maybe in Leicester, run by a small capitalist who may be an ex worker, takes orders from a couple of big brands that offer terms on a take-it-or-leave-it basis. The Marxist dichotomy cannot account for this pattern. Either the relation between the worker and his boss is the wage relation in which surplus value is created, which implies that the relation between the boss and the brand involves an exchange of equivalent values, or the relation between the boss and the worker is the true wage relation, but then the small boss should not be considered a capitalist, and we cannot explain how he is accumulating value. (Marx was somewhat conscious of these problems, and tried to solve them in volume III of Capital, but he had made it impossible for himself by locking in his theory of surplus value in the beginning of Volume I.)
If exploitation and surplus value creation happen both in the spheres of production circulation, the distinction itself is flawed. If production is defined as the whole process of formation of a use value ready for consumption, then the circulation of intermediate products between firms is a crucial part of production. And when we follow Marx into the realm of production, he actually analyzes the circulation of people and goods inside the workplace. The whole economy could alternatively be considered as one complex production process or one extended circulation network, with exploitation happening simultaneously at many different scales.
And there is no good reason to say that relations inside the workplace are relations of production, whereas relations between these units would be relations of circulation.
Which concept could then replace that of production as the central notion of a critical political economy? I offer an answer inspired by many political economists, notably the comparative theorist Janos Kornai and the law scholar Sanjukta Paul:coordination. To test that hypothesis, we need to confront the concept not only to commercial networks, but also to the traditional building blocks of capitalism : private property of the means of production, production for profit, commodification of labor power, and the power of capital inside the workplace. I argue that each of these fulfills, although partially and defectively, a coordination function. Private property of the means of production is a relation of mutual exclusion between owners, preventing them from the type of coordination failure known as the tragedy of commons. Patterns of profitability shape the flow of investment, therefore determining the allocation of the means of production across firms and industries. The existence of a labor market is a way to allocate labor power. i.e. to solve the coordination problem which Marx calls the ‘social division of labor’. Finally, the hierarchical power of capital inside the workplace is a way to manage the division of labor in manufacture, which is also a coordination problem. As Marx puts it, inside the workplace the capitalist fulfills the function of the conductor of an orchestra.
Exploitation and domination are embedded in the coordinating function of these relations. Private property consecrates the asymmetrical right of some to exclude others from the access to certain goods and services. Industrial profits are the archetypal form of exploitation in the Marxist sense. The existence of a labor market opens the possibility of unemployment. And the coordinating function of orchestra conductor is the very root of the despotism of capital eloquently described by Marx – a new version of the connection between coordination and vertical power which Hobbes theorized in its purest form.
By defining capitalism as a mode of coordination, we are therefore able to account both for the exploitation and domination implied by the building blocks of capitalism as traditionally defined, and for indirect, commercial or financial, forms of exploitation and domination. It can also account for aspects of capitalism which other critical schools of thought have analyzed and which Marxism has always been uneasy with, in particular the Keynesian problem of coordination between aggregate supply and aggregate demand, and the role played here by money, the medium of coordination par excellence in a market economy. Post-Keynesians have suggested both that this problem should be considered a structural characteristic of capitalism, and that it is intimately connected to the domination of the capitalist class.
What about the political implications of this conceptual setting? The first part of my answer is to remind that Marxism has always acknowledged that capitalism does fulfill some useful social functions, even if defectively. Capitalism does expand a society’s productive capacity for needed goods and services, as its apologists argue. But Marxism answers that the capitalist system of production is simultaneously a system of exploitation and domination, and that at some point of history it becomes possible to replace it by another mode of production, devoid of systematic exploitation.
Coordination has long been a theme dear to the liberal thinkers, in particular Hayek and Williamson, both of them defending the capitalist society in the name of coordination. These thinkers are right that capitalist institutions do fulfill important coordination functions, but they fail to track their defects, especially their exploitative aspects.
I think that three distinct political positions (with a continuum of possibilities in between) can be defined in this conceptual framework. The capitalist conservative position, that of Hayek and Williamson, holds capitalist coordination for unsurpassable – any institutional change directed against domination would lead to a loss in coordination, eventually prejudiciable to everyone. The capitalist progressive position, which seems to be the more or less conscious framework of the economics of inequality, argues that the coordination function of capitalist institutions can be maintained while removing their domination effects, e.g. by fiscal redistribution. A socialist position is also possible, that shares the conservative analysis of domination and coordination as two sides of the same coin–the capitalist mode of coordination–but believes in the possibility of reshaping our very coordination relations, to introduce new ones that would be as or more effective and be devoid of systemic domination.
The intellectual defense of the socialist position requires a careful theoretical and empirical study of alternative modes of coordination at the scale of the workplace, industries (public health and education sectors in many European countries), a national economy as a whole, and international economic relations. I am in no position to advance original positions on these matters, and I would just like to emphasize what we gain from framing the debates on socialism as debates on coordination. Marx and Engels conceived communism as a reconciliation of the social totality with itself, which the socialization tendency immanent to capitalism would help bring about. The logical consequence is that communism could and should do without law, which presupposes and reproduces the separation of society between different individuals and entities. In the view offered above, this is meaningless; the coordination problem is the essential economic problem and the so called immanent socialization was just a transition from some coordination relations to others. Society will remain a multiplicity of heterogenous parts, and it is up to transformative politics to arrange them in the best way.