The first four sections provide the theoretical basis for the concluding empirical descriptions, and if I'm not mistaken, the third and fourth sections are not entirely distinct. Further, Harvey argues that the first two sections are not necessary to reach the conclusions of the third, but I'm not sure myself. But at any rate, here are the steps Marx takes:
Section 1. The Increased Demand for Labour Power that Accompanies Accumulation, the Composition of Capital Remaining the SameIn the first section, Marx argues that the accumulation process eternally preserves the class relationship between workers and capitalists because of exploitation, the conversion of surplus-value into capital, wage labor, etc. Nothing new: "...the mechanism of the accumulation process itself not only increases the amount of capital but also the mass of the 'labouring poor,' i.e. the wage-labourers, who turn their labour-power into a force for increasing the valorization of the growing capital, and who are thereby compelled to make their relation of dependence on their own product, as personified in the capitalist, into an eternal relation (765, italics added).
Section 2. Relative Diminution of the Variable Part of Capital Simultaneously with the Progress of Accumulation and of the Concentration that Accompanies it
Section 3. Progressive Production of a Relative Surplus Population or Industrial Reserve Army
Section 4. Different Forms of the Relative Surplus Population. The General Law of Capitalistic Accumulation
Next, Marx distinguishes between the concepts of concentration and centralization, which he defines in a bit of a counter-intuitive manner. Concentration is merely the fact of extended accumulation by one company in isolation, like a spiral that spins outwards but with no regard for others. But eventually, these growing companies will collide with one another, and they can't simply keep accumulating infinitely without some companies being eaten and others eating them. Centralization is Marx's term for this. It describes a process wherein capitals are attracted to one another, where capital grows not by simply creating wealth ex nihilo (that is, from workers or from credit) but rather by eating the capital of others. So which capital b/eats which capital? Generally the one able to steal all the business by selling cheaper commodities, that is, the capital with greater productivity. And hence "this depends in turn on the scale of production. Therefore the larger capitals beat the smaller" (777). As a general rule, contra the self-regulating dream of supply and demand, the free market breeds unfair competition and monopolies.
Third, Marx pays attention to the drive for productivity , resulting in more capital invested in machines which obviate the need for labor. In other words, variable capital goes down, constant capital goes up. The contradiction of course is that, with competition, the exchange-value of these commodities will soon be canceled out by the technical advances of other capitals, even as use-values are increased; meantime, the rate of surplus-value has decreased because the source of surplus-value, that is, exploited labor, has been decreased relative to the total mass of capital invested. In other words, profit will fall as use-values rise.
Harvey has a good exposition on whether or not this falling profit is a law, a tendency, a tendency of law, a law of tendency, etc. It's worth listening to, around the 43:30 mark and again at the 81:40 mark:
It's also worth checking out a summary of crisis theories from Anwar Shaikh, an economist at the New School, one of the better to have written on this (for a better bibliography, read Harvey's Limits to Capital and search for "falling rate of profit" in the index).
Harvey takes a pretty good position on the truthiness of Marx's law: the contradiction of technology that displaces labor as the source of exploitation is always a disruptive force in capitalist development, and Marx's descriptions of this process are accurate, but they are not predetermined as an economic law insofar as, well, even though profit has fallen throughout history, people have found new ways to generate more profit, and there is still profit to be made today (by the way, I'm going to resist the temptation to write about the massive fuckup that is the current financial crisis, but the connections between the present day and Marx's thoughts on credit, finance capital and deregulation are readily evident).
Not only does capital shoot itself in the foot by displacing labor, Harvey concludes, but it also generates a condition that is worse for the worker. When there is more supply of labor than there is demand, workers' wages decrease, and more exploitation can be squeezed out of employed workers who are expendable and superfluous to the actual production process. De-skilling and machinery are a big part of this. And of course, one of the lessons of neoliberalism has been that when the inevitable crises do occur, the capitalist classes will try generate more capital, that is, more surplus-value, by squeezing wages down even further; and they get away with it because of the pressure exerted by the industrial reserve army in the backdrop ("Sure you can quit, but I'm just going to hire that guy instead when you leave").
Marx thus ends on the most crushing note of the capitalist process, that is, the way in which it produces and fixes a relation between superfluous working classes and the capital they have become dependent upon, even as the capital displaces them from any sort of political, economic security. In my mind, I couldn't help think of the somewhat trendy yet completely de-historicized attention paid recently to such populations, given various names by various theorists: Agamben and homo sacer; Foucault and biopolitics; Subaltern studies; the various decontextualized others of Levinas and other literary critics; etc. etc. Isn't it necessary to re-assert the history of how such populations are reproduced by particular socio-economic dynamics? Isn't it incumbent to demonstrate the underlying, broadly encompassing processes that unify these snapshots scattered across and time and place? Unfortunately, it doesn't seem to happen much.
My final thought on this chapter and on Marx's argument as a whole is that it seems very difficult to pin down precisely a single concept the process that holds the key to its effectiveness. This is perhaps why the left has been fighting with itself for so long over what to do about capital. It's not just about exchange-value, it's not just about machines or money, it's not just about wages or time, although all of those factors are very much integral to the history of the modern world.
What really strikes me about Marx's presentation is that capital's strongest weapon is that it grows. When it grows, it displaces people who are forced to rely upon capital to survive, and the more it grows, not only do the alternatives disappear but more pressure is exerted upon the worker and the capitalist classes to adhere ever more stringently to capital's laws. Capital becomes personified.
In the case of wages, workers' wages will only go down once it becomes clear that alternatives for non-wage based labor have disappeared. In the case of the organic composition of capital (variable and constant capital ratios), only when industries fight one another in such a deadlock will the pressure for technological changes enabling greater productivity actually generate faster, better, bigger machines.
If there were a name for this dynamic, it would be the dynamic of competition. But competition is an empty term, a formal one that lacks any actual commentary on who is competing with whom, what draws them together, who wins and why they win. Again, in order to explain why it works, one must start explaining the different, interacting levers of action that spur accumulation, centralization and crises.
At some point, Marx will make arguments about capital that is not yet fully mature, stages in production that cannot yet generate systemic crises or throw laborers into a state of floating insecurity. For example, on 785:
This peculiar cyclical path of modern industry, which occurs in no earlier period of human history, was also impossible when capitalist production was in its infancy. The composition of capital at that time underwent only very gradual changes . . . . Even though the advance of accumulation was slow in comparison with that of the modern epoch, it came up against a natural barrier in the shape of the exploitable working population; this barrier could only be swept away by the violent means we shall discuss later.In a footnote inserted into the French edition on the next page:
But only after mechanical industry had struck root so deeply that it exerted a preponderant influence on the whole of national production; only after foreign trade began to predominate over internal trade, thanks to mechanical industry; only after the world market had successively annexed extensive areas of the New World, Asia and Australia; and finally, only after a sufficient number of industrial nations had entered the arena -- only after all this had happened can one date the repeated self-perpetuating cycles, whose successive phases embrace years, and always culminate in a general crisis, which is the end of one cycle and the starting-point of another (786).These paragraphs suggest to me that capital's force lies in its imperative for growth, which comes not from a single source but rather from the mutual reinforcement from competition among many different sources. This is more than simply saying it cancels out other alternatives; it makes any decision to seek an alternative irrational and suicidal. Even if competition destroys itself, the stakes are too high to allow that destruction to occur, for capital is too mutually dependent to simply allow anyone to leave, cough, bailout (sorry). Competition, more than simply fixing a relationship of exploitation or drawing a line between the haves and have nots, regulates that line and manipulates it in order to maximally intensify exploitation to its logical limit. Much like the self-regulating supercomputers behind hedge fund schemes, capital is less about subjective and more about objective limits and forces. Even if it is regulated, that regulation, definitionally, only aids and abets that extremism. And it ends in scenes as ugly as this:
Capital acts on both sides at once. If its accumulation on the one hand increases the demand for labour, it increases on the other the supply of workers by 'setting them free,' while at the same time the pressure of the unemployed compels those who are employed to furnish more labour, and therefore makes the supply of labour to a certain extent independent of the supply of workers. The movement of the law of supply and demand of labour on this basis completes the despotism of capital (793).