Originally titled “Marx on the Corporation.” Republished with permission from the author. Consider reading Mason’s short “Notes on Capital” first, to give the discussion below its proper context. 
Let’s begin at the beginning.
Capital, for Marx, is not a thing, it’s a social relation, a way of organizing human activity.  Or from another point of view, it’s a process. It’s the conversion of a sum of money into a mass of commodities, which are transformed through a production process into a different mass of commodities, which are converted back into a (hopefully greater) sum of money, allowing the process to start again. Capital is a sum of money yielding a return, and it is a mass of commodities used in production, and it is a form of authority over the production process, each in turn.
When we have a single representative enterprise, managed by its owner and financed out of its own retained profits, then there’s no need to worry about where the “capitalist” is in this process. They are the owner of the money, and they are the steward of the means of production, and they are master of the production process. Whatever happens in the circuit of capital, the capitalist is the one who makes it happen.
This is the framework of Volume 1 of Capital. There the capitalist is just the personification of capital. But once credit markets allow capitalists to use loaned funds rather than their own, and even more once we have joint-stock enterprises with salaried managers in charge of the production process, these roles are no longer played by the same individuals. And it is not at all obvious what the relationships are between them, or which of them should be considered the capitalist. This is the subject of part V of Volume 3 of Capital, which explores the relation of ownership of money as such (“interest-bearing capital”) with ownership of capitalist enterprises.
For present purposes, the interesting part begins in chapter 23. There Marx introduces the distinction between the money-capitalist who owns money but does not manage the production process, and the industrial, functioning or productive capitalist who controls the enterprise but depends on money acquired from elsewhere. “The productive capitalist who operates on borrowed funds,” he writes, “represents capital only as functioning capital,” that is, only in the production process itself. “He is the personification of capital as long as […] it is profitably invested in industry or commerce, and such operations are undertaken with it […] as are prescribed by the branch of industry concerned.”
The possibility of carrying out a capitalist enterprise with borrowed funds implies a division of the surplus into two parts — one attributable to management of the enterprise, the other to ownership as such. “The specific social attribute of capital under capitalist production — that of being property commanding the labour-power of another” now appears as interest, the return simply on owning money. So “the other part of surplus-value — profit of enterprise — must necessarily appear as coming not from capital as such, but from the process of production […]. Therefore, the industrial capitalist, as distinct from the owner of capital [appears] […] as a functionary irrespective of capital, […] indeed as a wage-labourer.”
So now we have one set of individuals personifying capital at the
M moment, when capital is in its most abstract form as money, and a different set of individuals personifying it in the
P moments, when capital is crystallized in a particular productive activity. 
One effect of this separation is to obscure the link between profit and the labor process:
The money-owners who receive profit in the form of interest (or dividends) are different from the actual managers of the production process.
Not only that, the two often experience themselves as opposed.
In this sense, the division between the money-capitalist and the industrial capitalist blurs the lines of social conflict.
Interest as such expresses […] the ownership of capital as a means of appropriating the products of the labour of others. But it represents this characteristic of capital as something which belongs to it outside the production process […]. Interest represents this characteristic not as directly counterposed to labour, but rather as unrelated to labour, and simply as a relationship of one capitalist to another. […] In interest, therefore, in that specific form of profit in which the antithetical character of capital assumes an independent form, this is done in such a way that the antithesis is completely obliterated and abstracted. Interest is a relationship between two capitalists, not between capitalist and labourer.
We might read Marx here as warning against an easy opposition between “productive” and “financial” capital, in which we can with good conscience take the side of the former. On the contrary, these are just shares of the same surplus extracted from us in the labor process. It’s important to note in this context that Marx speaks of a “productive capitalist,” not of productive capital. The productive capitalist and the money capitalist are, so to speak, two human bodies that the same capital occupies in turn.
Once the pirates have burned your fields, seized your possessions and carried off your daughters, it shouldn’t matter to you how they divide up the booty: I think this is a valid reading of Marx’s argument here. Or as he puts it: “If the capitalist is the owner of the capital on which he operates, he pockets the whole surplus-value. It is absolutely immaterial to the labourer whether the capitalist does this, or whether he has to pay a part of it to a third person as its legal proprietor.”
But while the development of interest-bearing capital obscures the true relations of production in one sense, it clarifies them in another. It separates the claims exercised by ownership as such, from the claims due to the specific labor performed by the capitalist within the enterprise. With the owner-manager, these two are mixed together. (This is still a big problem for the national accounts.)  Now, the part of apparent profit that was really payment for the labor of the capitalist appears in a distinct form as “wages of superintendence.”
Marx’s analysis here seems like a good starting point for discussions of the position of managers in modern economies.
The specific functions which the capitalist as such has to perform, […] [with the development of credit] are presented as mere functions of labour. He creates surplus-value not because he works as a capitalist, but because he also works, regardless of his capacity of capitalist. This portion of surplus-value is thus no longer surplus-value, but its opposite, an equivalent for labour performed. […] [T]he process of exploitation itself appears as a simple labour-process in which the functioning capitalist merely performs a different kind of labour than the labourer.
As Marx later emphasizes, one consequence of the development of management as a distinct category of labor is that the profits still received by owners can no longer be justified as the compensation for organizing the production process. But what about the managers themselves, how should we think about them? Are they really laborers, or capitalists? Well, both — their position is ambiguous. On the one hand, they are performing a social coordination function, that any extended division of labor will require. But on the other hand, they are the representatives of the capitalist class in the coercive, adversarial labor process that is specific to capitalism.
The discussion is worth quoting at length:
The labour of supervision and management is naturally required wherever the direct process of production assumes the form of a combined social process, and not of the isolated labour of independent producers. However, it has a double nature. On the one hand, all labour in which many individuals co-operate necessarily requires a commanding will to co-ordinate and unify the process […] much like that of an orchestra conductor. This is a productive job, which must be performed in every combined mode of production.
On the other hand […] supervision work necessarily arises in all modes of production based on the antithesis between the labourer, as the direct producer, and the owner of the means of production. The greater this antagonism, the greater the role played by supervision. Hence it reaches its peak in the slave system. But it is indispensable also in the capitalist mode of production, since the production process in it is simultaneously a process by which the capitalist consumes labour-power. Just as in despotic states, supervision and all-round interference by the government involves both the performance of common activities arising from the nature of all communities, and the specific functions arising from the antithesis between the government and the mass of the people.
In one of those acid asides that makes him so bracing to read, Marx quotes an American defender of slavery explaining that since slaves were unwilling to do plantation labor on their own, it was only right to compensate the masters for the effort required to compel them to work. In this sense it doesn’t matter that the Bosses are performing productive labor. Their claims are just a version of the German nihilists’ “It’s only fair that you give me what I want, since I’ve gone to such effort to take it from you.”  Or Dinesh D’Souza’s argument that equality of opportunity would be unfair to him, since he’s gone to great effort to give his kids an advantage over others. 
But again, the industrial capitalist is not only a slave-driver. They do have an essential coordinating function, even if it is performed by the same people, and in the same activities, as the coercive labor-discipline that extracts greater effort from workers and deprives them of their autonomy. The ways these two sides of the labor process develop together is one of the major contributions of Marxist and Marx-influenced work, I think — Braverman, Noble, Marglin, Barbara Garson. It seems to me that, paradoxical as it might sound, it’s this positive role of managers that is ultimately the stronger argument against capitalism. Because the development of professional management fatally undermines the supposed connection between the economic function performed by capitalists, and the economic form of property ownership.
Marx makes just this argument:
The capitalist mode of production has brought matters to a point where the work of supervision, entirely divorced from the ownership of capital, is always readily obtainable. It has, therefore, come to be useless for the capitalist to perform it himself. An orchestra conductor need not own the instruments of his orchestra, nor is it within the scope of his duties as conductor to have anything to do with the “wages” of the other musicians. Co-operative factories furnish proof that the capitalist has become no less redundant as a functionary in production […]. Inasmuch as the capitalist’s work does not […] confine itself solely to the function of exploiting the labour of others; inasmuch as it therefore originates from the social form of the labour-process, from combination and co-operation of many in pursuance of a common result, it is […] independent of capital.
The connection Marx makes between joint-stock companies (what we would today call corporations) and cooperative enterprises is to me one of the most interesting parts of this whole section. In both, the critical thing is that the work of management, or coordination, is just one kind of labor among others, and has no necesssary connection to ownership claims.
The wages of management both for the commercial and industrial manager are completely isolated from the profits of enterprise in the co-operative factories of labourers, as well as in capitalist stock companies. […] Stock companies in general — developed with the credit system — have an increasing tendency to separate this work of management as a function from the ownership of capital […] just as the development of bourgeois society witnessed a separation of the functions of judges and administrators from land-ownership, whose attributes they were in feudal times. Since, on the one hand, […] money-capital itself assumes a social character with the advance of credit, being concentrated in banks and loaned out by them instead of its original owners, and since, on the other hand, the mere manager who has no title whatever to the capital, […] performs all the real functions pertaining to the functioning capitalist as such, only the functionary remains and the capitalist disappears as superfluous from the production process.
This, to me, is one of the central ways in which we can see capitalism as a necessary step on the way to socialism. Only under capitalism has large scale industry developed; only the acid of the market was able to break the bonds of small family productive units and free their constituent pieces for recombination on a much larger scale. So the only form in which the organization of large-scale enterprises is familiar to us is as capitalist enterprises. (At least, this is Marx’s argument. Arguably he understates the ability of states to organize production on a large scale.) But just because large industrial enterprises and capitalism have gone together historically, it doesn’t follow that that capitalism is the only institutional setting in which they can exist, or that the conditions required for their development are required for their continued existence.
In fact, as capitalist enterprises develop, their internal organization becomes progressively less market-like. Markets exist only at the surfaces, the external membranes, of enterprises, which internally are organized on quite different principles; and as the scale of enterprises grows, less and less economic life takes place on those surfaces. So while capital continues, nominally, to be privately owned, relations of ownership play less and less of a role in the concrete organization of production. The “mere manager” as Marx says, “has no title whatever to the capital”; nonetheless, he or she “performs all the real functions” of the capitalist.
When Marx was writing this in the 1870s, he thought the trend towards the separation of ownership from control was clearly established, even if most capitalist enterprises at the time were still directly managed by their owners.
With the development of co-operation on the part of the labourers, and of stock enterprises on the part of the bourgeoisie, even the last pretext for the confusion of profit of enterprise and wages of management was removed, and profit appeared also in practice as it undeniably appeared in theory, as mere surplus-value, a value for which no equivalent was paid.
That’s as far as the argument gets in chapter 23.
The next few chapters are focused on the other side of the question, interest-bearing capital — that is, capital that appears to its owners simply as money, without being embodied in any production process. Chapter 24 is an attack on writers who reduce both to money capital, and imagine that the accumulation of capital is just an example of the power of compound interest. (Among other things, this chapter anticipates the essential points of left critiques of Piketty by people like Galbraith and Varoufakis, and by me.)  Chapter 26 attacks the opposite conflation — the treatment of money as just capital in general, and of interest as simply a reflection of the physical productivity of capital rather than a specifically monetary phenomenon. This is today’s orthodoxy, represented for Marx by Lord Overstone. Chapter 25 anticipates Minsky on the elasticity of finance, and takes the side of the credit-money theorists like Thornton and banking-school writers like Tooke and Fullarton, against quantity theorists and the currency school. Marx’s debt to Ricardo is well known, but it’s less recognized how much he learned from this group of writers — the best discussion I know is by Arie Arnon.  When Tooke died, Marx wrote to Engels that he had been “the last English economist of any value.”
Marx returns to the industrial or functioning capitalist in chapter 27, which is focused on joint-stock companies. Marx credits stock companies with “an enormous expansion of the scale of production and of enterprises, that was impossible for individual capitals.” And critically these new enterprises are public in both name and substance (the “public” in “publicly-traded corporations” is significant).
The development of joint stock companies continues the sociological transformation that begins with the development of interest-bearing capital and the ability to operate on borrowed funds — that is, the
transformation of the actually functioning capitalist into a mere manager, administrator of other people’s capital, and of the owner of capital into a mere owner, a mere money-capitalist. Even if the dividends which they receive include the interest and the profit of enterprise, […] this total profit is henceforth received only in the form of interest, i.e., as mere compensation for owning capital that now is entirely divorced from the function in the actual process of reproduction, just as this function in the person of the manager is divorced from ownership of capital. […] This result of the ultimate development of capitalist production is a necessary transitional phase towards the reconversion of capital into the property of producers, although no longer as the private property of the individual producers, but rather […] as outright social property. […] [T]he stock company is a transition toward the conversion of all functions in the reproduction process which still remain linked with capitalist property, into mere functions of associated producers.
In short, the joint stock company “is the abolition of the capitalist mode of production within the capitalist mode of production itself.”