Marxists and those influenced by Marxist analyses of capitalism, to greater or lesser degree, have been unusually reticent about theorizing demographic change. Demography is a specialized branch of social science to which Marxist and marxisant contributions have been largely absent although there are often passing references to demographic change.² I return to explore aspects of this silence after first revisiting Marx on population.

I will not dwell on Marx's savage critique of Malthus who explained “‘over-population’ by the eternal laws of nature, rather than the merely historical laws of the nature of capitalist production” (Marx 1976: 666, note 7). I focus instead on Marx's notion of “surplus population”:

It is capitalist accumulation itself that constantly produces, and produces indeed in direct relation with its own energy and extent, a relatively redundant working population, i.e. a population which is superfluous to capital's average requirements for its own valorization, and is therefore a surplus population. . . a law of population peculiar to the capitalist mode of production; and in fact every particular historical mode of production has its own special law of population, which are historically valid within that particular sphere. (Marx 1976: 782–784)

This “law of supply and demand of labor” (a term used by Marx) regulates “the general movement of wages or the ratio between the working class—i.e., the total sum of labour-power—and the total social capital” (Marx 1976: 790–791, emphasis added). Its principal mechanism is the inbuilt drive of capital to develop the productive forces: “The higher the productivity of labour, the greater is the pressure of the workers on the means of employment, the more precarious therefore becomes the condition for their own existence” (ibid.: 799, emphasis added); “machinery produces a surplus working population” (ibid.: 531–532).

Is the impetus toward “surplus population” an effect of capital's law of accumulation or a “function” of it?:

If a surplus population of workers is a necessary product of accumulation. . .it becomes a condition for the existence of the capitalist mode of production. . . . Effects become causes in their turn, and the various vicissitudes of the whole process, which always reproduces its own conditions, take on the form of periodicity. . . the constant reproduction of a relative surplus population of workers is a necessity of capitalist accumulation. (Marx 1976: 784, 786–787, emphases added)

Do these statements—notably “effects become causes”—point toward a “functional” view of “surplus population” (one with problematic legacies)? Wisely Marx did not attempt any general elaboration of the extent to which a “surplus population”/reserve army of labor arises, of what size and scale, nor when, how and why. At the same time, he notes that “like all other laws, it is modified in its working by many circumstances” (Marx 1976: 798). It is those “many circumstances” (historical specificities), of course, which are of foremost concern to those undertaking historical and ethnographic research, and which enter into many of the illustrations Marx provided in Capital Volume 1, Chapter 25.

In the tension between surplus population as effect, albeit systemic effect, and function of the dynamics of capitalist accumulation, I prefer the former, especially in any investigation of its many and diverse concrete circumstances. I do not know of any arguments about the peripheries of the capitalist world economy that center on the (relentless) development of the productive forces there, whatever other benefits (or “functions”) peripheral “surplus populations” are suggested to have for world capitalism. Rather such “benefits”/”functions” are typically framed in terms of either the “cheapness” of labor power or its conditions of reproduction and wage levels, or both.

Another point I would make here concerns the “supply” side of the “law of supply and demand of labor”:

Capitalist production can by no means content itself with the quantity of disposable labour power which the natural increase of population yields. It requires for its unrestricted activity an industrial reserve army which is independent of these natural limits. (Marx 1976: 788, emphases added)

Apart from specific instances given by Marx of dispossession (not least of agricultural labor) to illustrate concrete historical circumstances of the supply of (industrial) workers, and observations about their gender and age composition, there is no theorization of the demographic sources of surplus population. Indeed, apart from the polemic against Malthus (and others), there is no “theory of population” in Marx nor any formulation of a “special law of population” peculiar to the capitalist mode of production. How serious is this?

At the heart of this particular matter is that labor power, at the core of Marx's theory of value, is not produced as a commodity. As Wally Seccombe (1993: 9) put it: “Labour-power is consumed in the workplace; at home it is produced apart from capital”; “The family-formation strategies of subordinate classes respond to their own subsistence imperatives; they must never be treated as if producers, by balancing the dependency ratios of their households, were striving to reproduce the present mode of production or to satisfy the demands of their masters” (ibid.: 3).

Advances along the lines indicated by Seccombe are registered in his own extraordinary investigations of demographic change in transitions from feudalism to capitalism in northwestern Europe (1992) and in the European industrial revolutions of the eighteenth to early twentieth centuries (1993). Also, of course, major advances are registered in materialist feminist work on the dynamics and domains of social reproduction (e.g., Bhattacharya 2017). However, these advances still fall short of tackling a “law of population” peculiar to capitalism (Seccombe 1983, 1993), leaving us with the residual of Marx's “natural” growth of and limits to population (aforementioned). There are probably two major reasons for this lacuna. One is the antipathy, inherited from Marx, to the Malthusian legacy and the deeply reactionary purposes it continues to serve. At the same time, it can be argued that Malthus's questions, suitably reformulated, are still important even if his answers are rejected (Martínez-Alier 2002). The other reason, much less remarked, is the intellectual problem, perhaps intrinsic to demography, of inferring the reasons for individual “choices” or “decisions” about (and impositions of) fertility practices from aggregated patterns of population change, as in standard models of “demographic transition” based in Western history.³ For Marxists, of course, individual fertility “decisions” (“choices”) are structured by the class and gender patterns of “subsistence imperatives”, or pressures on reproduction, as Seccombe (previously quoted) rightly emphasized. Two examples of explaining reductions in fertility and persistent high fertility by the conditions of existence of different classes are, concerning the former, Seccombe's own work on dynamics of working-class family change in Europe's “demographic transition” (Seccombe 1993); and, concerning the latter, Mamdani's investigation of poor peasants in Uganda who exploit the only resources available to them, namely labor on their own small farms (with inferior hand tools), and their powers of procreation (Mamdani 1987; see also Mamdani 1974).

Turning to arguments about “surplus populations” in the peripheries of global capitalism (the Third World), there are two approaches worth remarking on here. One was tabled early on, as far as I know, in Latin American debates about “marginality” (e.g., Obregón 1974, among the first to appear in English translation) in which “the marginal poor were deemed to be ‘a-functional’ to the needs of monopoly capitalism, unlike the classic ‘reserve army of labour’ analysed by Marx for an earlier era”—in short, a much more expansive conception of the “surplus” in “surplus population” (Munck 2013: 748).⁴ Ronaldo Munck (2013) also usefully traces and criticizes subsequent iterations of “marginality”/exclusion: “informality”, “precarity” and the like. His paper was published before a further and current notion of “surplus population” in the work of James Ferguson (2015), Tania Murray Li (2017) and Ferguson and Li (2018), among others. For them, as for some of the Latin American “marginalists”, any prospects of a “proper job” for a vast and increasing number of the world's poor have vanished in the world of twenty-first century capitalism to which they are largely superfluous, especially but not exclusively in its peripheries—thus excessive supply in terms of any “law of supply and demand of labour” (Ferguson and Li 2018).⁵ Ferguson and Li's principal interest is to explore the many diverse ways in which the poor are able to construct their livelihoods, and the social resources they draw on to do so, and to argue for a new politics of redistribution radically different from that which prevails in capitalism today, but of which they believe there are hopeful experiments like Basic Income Grants irrespective of “gainful” (wage) employment.

In a somewhat similar vein concerning the “excess” supply of (potential) workers, Mike Davis proposes the following:

[An] outcast proletariat. . .the fastest growing and most novel social class on the planet. . .[this] urban informal working class. . .[is] not a labor reserve army in the nineteenth-century sense. . .[but] a mass of humanity structurally and biologically redundant to global accumulation. . .the new urban poor are. . .a surplus humanity [who] will [somehow] exist outside the formal relations of production. . .the labor-power. . .expelled from the formal world economy. (Davis 2004: 11, 13–14, emphases added).

If the use of “surplus population” so briefly outlined seems to connect more with what I term the effects of capitalist development, a second approach veers more toward its functional interpretation. This is that ostensibly “surplus population(s)” in the peripheries are not only produced as an effect of their incorporation in world capitalism but are constitutive of its functioning, a position associated with various iterations of “the development of underdevelopment”, dependency theories and the like. In short, the “backwardness”, “incompleteness”, or whatever, of capitalist development in the South, together with its “surplus population(s)” is a necessary condition of capitalist accumulation in the North, hence underwrites capitalist accumulation on its global scale.

The difference between the two perspectives is not only theoretical (and political) but also empirical. For Munck (2013) among others—and in direct opposition to Ferguson and Li, and Davis, for example—there is “massive expansion of the global working class in classic Marxist forms” (Munck 2013: 754), and the “basic fact is that long-term employment is rising” (ibid.: 753), while at the same time “a key differentiator” between South and North remains the continuing size of the “informal proletariat” in the former (ibid.: 755). The logic of the argument here is that of Marx on “surplus population” and the “industrial reserve army” as necessary and central to the accumulation of capital, albeit similarly silent on the “supply” side of any “law of supply and demand of labor”.

It is worth making a slight digression at this point about the meanings assigned to the dynamics of “proletarianization”. There are two key moments in the processes of proletarianization. The first is that of dispossession of the means of production, hence reproduction, which registers the availability of “free” wage labor in Marx's ironic formulation, creating a class compelled to work for the possessors of means of production (capital) under the “dull compulsion of economic relations” (Marx 1976: 899). This resonates in Michael Denning's insistence “that ‘proletarian’ is not a synonym for ‘wage labourer’ but for dispossession, expropriation and radical dependence on the market. You don't need to be a proletarian: wageless life, not wage labour, is the starting point in understanding the free market” (Denning 2010: 81, emphasis added). The second moment is the entry into wage work to produce surplus value for capital. The impact of the first moment without realization of the second is neatly encapsulated in Deepankar Basu and Debarshi Das's formulation (2009: 158) of “dispossession without proletarianization”.

Notions of “surplus population” and the reserve army of (industrial) labor tend to cluster around the first of these two moments, but more generally they point toward the variable connections, spaces and gaps, between them. Moreover, the “excess” of those created by the first moment but not absorbed into the second is where disagreements about the scale of “surplus population” and its key characteristics—notably its “proletarian” or “semi-proletarian” nature or otherwise—lurk (as just illustrated). This suggests the need for further determinations, even at this level of generality, for example, in the context of debates about “informality”.

A particular element in intense and far-ranging debates about the “informal” sector/economy/labor (and even “informal proletariat”, as Munck and Davis, aforementioned) that deserves comment concerns petty commodity production, both agrarian and non-agrarian, rural and urban. Petty commodity production, in all its intrinsic diversity of forms and locations in social divisions of labor, so often enters the reproduction/livelihood practices of what I prefer to call “classes of labor” in both North and South—prefer, that is, to essentialized notions of “proletariat” (in much of the Marxist tradition, and hinted at by Denning quoted earlier) or “peasantry” (in the Chayanovian tradition).⁶ Simply, “classes of labor” refers to “the growing numbers. . .who now depend—directly and indirectly—on the sale of their labour power for their own daily reproduction” (Panitch et al. 2000: ix, emphasis added), echoed by Aaron Benanav (2019: 680, emphasis added): the (rising) share of “the population that depends on selling its labor—or the simple products of its labor—to survive”.

It is not adequate to dismiss conceptions of “informal economy” and petty commodity production as obscuring an ostensible reality of comprehensive proletarianization as (more or less conventional) wage employment as, for example, Saumyajit Bhattacharya (2014) does in his critique of Barbara Harriss-White (focusing on Harriss-White 2010, among her many important essays on this subject). The fundamental condition of existence of petty commodity production today is the social relation between capital and (wage) labor in generalized commodity production (Bernstein 2010, Chapter 7). While in some instances much of the labor in farming, as well as in urban and rural “informal economy”, might be “disguised wage labor”, this does not contradict the widespread prevalence— as well as diversity and complexity—of petty commodity production in contemporary capitalism, which contains both “petty accumulation” and “sub-subsistence” activity at its poles (in Davis's terms) and into which one particular insight is given by considerations of “accumulation from below” (for example, in the lucid essay by Hall 2012, and the excellent ethnography of Li 2014). In short, debates about petty commodity production often illuminate consideration of class dynamics within “informal economy”, and of (peripheral) “classes of labor” more generally in contemporary capitalism, rather than obscure them pace S. Bhattacharya.

“Functional(ist)” notions of “surplus population” are often associated with arguments about “primitive accumulation” as “inherent-continuous” (in De Angelis's term, 2007) throughout the history of capitalism, in a line of descent from Rosa Luxemburg on imperialism (1951) to David Harvey's notion of “accumulation through dispossession” as capital's response (“fix”) to a long period of declining profitability (2003). Invocations of (ongoing) “primitive accumulation” have enjoyed a widespread revival in recent decades in often promiscuous forms. A strong statement of this position is Sam Moyo et al. (2013: 186): “exploitation under monopoly capitalism has assumed starkly different forms in centre and periphery, up to the present, with the weight of primitive accumulation falling largely on the latter”. This is manifested in “permanent semi-proletarianization”:

On a global scale, we may affirm that capitalism has always sought to create the conditions for the perpetuation of non-remunerated labour outside the market and the displacement of the cost of social reproduction onto the labourers themselves. . .the expulsion of small producers from the countryside without their full absorption into the industrial or service sectors, or their permanent urbanization. This expelled population has a fundamental function in the world economy, not merely as a labour reserve but. . .which also ‘subsidizes’ the reproduction of capital by its own unremunerated labour. The self-exploitation of the semi-proletariat is a key dimension of super-exploitation, and is itself an extra-economic contribution to capital, in the sense of not being accounted for by the market. (Moyo et al. 2013: 187)

One observation about such formulations is that they are supported by no systematic demographic accounting or theses, especially regarding the “supply” side of “the law of supply and demand for labor”.⁷ Another issue, with crucial analytical implications, is the final statement quoted of the “non-market” qualities of “semi-proletarian” “self-” or “super-exploitation” as an “extra-economic contribution to capital”. This is striking as for Moyo et al. (2013) the “underdevelopment” of the South does not require it being “outside” of capitalism (in Luxemburg's sense) but is deemed to manifest “primitive accumulation” by virtue of its “non-remunerated labor” outside the market, and “extra-economic” mechanisms within world capitalism, through which “super-exploitation” occurs. In short, there is a conflation at work between, on one hand, the “inside” and “outside” of capitalism, and, on the other hand, the “economic” and “extra-economic” (or “non-economic”).

It is not clear what “outside the market” signifies theoretically in such statements, as distinct from being “outside” capitalism in the Luxemburgian tradition. Rather, it seems to me that “primitive accumulation” serves as a great residual of anything that was not tabled and “resolved” in Capital, and is thrown up by the histories of capitalism in the past 150 years, say. Is the answer to be found in notions of “primitive accumulation” deploying mechanisms outside “capitalism proper” while “functional” to capitalist accumulation? Why is “primitive accumulation” currently so popular? And what might be an alternative approach to some of the questions it raises?

There is a profound irony at work here. That is, many uses of continuing “primitive accumulation” (with all it is held to explain) rest on a forced (“stereotypical” in Lenin's term) binary between what is held to be capitalism “proper” and relations, forms and practices within the worlds of capitalism that deviate from it. It follows that the stricter and narrower the conception of the former (capitalism “proper”), the wider the scope for highlighting the latter (“inherent-continuous” primitive accumulation). The irony is compounded by assimilating (“mainstream”) “functionalist” readings of Capital to explanations of continuing “primitive accumulation” as more or less purposefully designed by the logic, interests or needs of capital in general—a common form of capital-centric reasoning. A further aspect of the irony suggested is that relatively few(er) Marxist thinkers and political formations adhere, at least explicitly, to (“orthodox”) conceptions of capitalist development today as “progress”.

There is no reason why that fundamental aspect of the (necessarily abstract) theory of the capitalist mode of production—the reproduction of labor through the “dull compulsion of economic relations” without the (regular) use of force—should imply that the concrete trajectories of accumulation to the present day are unmarked by coercion and violence or that there is no “political” that permeates the “economic” processes of capitalism, that the theory of the mode of production does not require attention to the capitalist state (or states). Or that histories of accumulation—in all their contradictions, specificities, different patterns, rhythms, fluctuations, violence, and outcomes—do not require analysis of the formation, functioning and activities of states, of the uneven course of class and popular struggles. This is a bizarre conclusion to attribute to Marx (and to Marxism), and is the effect, surely, of applying—in “stereotypical” fashion—the abstractions of Capital directly to the terrain of empirical/concrete analysis without mediation (“concentration of many determinations” in Marx's term).

Pointing toward an alternative way of using Marx, and doing “theory as history” in all its complexity, is Jairus Banaji's argument that capitalist relations of production cannot be read off from any specific form of the immediate process of production, specific form of exploitation, or type of enterprise, but are constituted by the laws of motion of the mode of production as a whole; “the immediate process of production can be structured in all sorts of ways, even under capitalism” (2010: 4). More generally, Banaji states the following:

Relations of production are simply not reducible to forms of exploitation, both because modes of production embrace a wider range of relationships than those in their immediate process of production and because the deployment of labour, the organisation and control of the labour-process, ‘correlates’ with historical relations of production in complex ways. (ibid.: 2010: 41).

This is a liberating perspective that opens up new ways of looking at the range of forms of production, and of social relations and practices, within an expansive understanding of capitalism even if inevitably it brings other issues in its train and (so far) leaves largely unanswered difficult questions about periodizing the emergence and development of capitalism's “laws of motion” on a world scale (Bernstein 2013). I do not write a blank check for Banaji's brilliant and often idiosyncratic work, despite its enormous stimulation; like any other, its method must be put to the test of concrete investigation and analysis which he has undertaken in relation to different historical contexts, before and since the emergence of world capitalism. Nonetheless, “theory as history” provides a point of departure very different from a “stereotypical” view of what capitalism in all its constituent social and spatial sites and dynamics “should” look like.

Banaji's perspective is not a theoretical panacea (does any such thing exist?) but a starting point for exploring whether the myriad diversity of forms of social relations, and of production, of labor regimes, and the like, in capitalism are best investigated through developing core analytical concepts and methods of Capital, albeit not to the exclusion of other determinations (which Marx accepted). Further, as Andreas Malm (2017: 196) remarks: “The proceedings of historical materialism are rich enough to contain other resources, beyond what Marx himself wrote”.

My own addition to this perspective is to suggest that primitive accumulation marked a period, or era, of world capitalism, that came to an end when the generalization of commodity relations was completed. That is to say, when the reproduction of any class, social category or formation, became impossible outside capitalist commodity relations, even if reproduction is not constituted exclusively by them. I have proposed that this era was broadly completed by the end of the colonial period in Asia and Africa, while recognizing, first, the unevenness of this process and, second, that the diversity of its modalities and forms was not simply the outcome of colonial design, which indeed often thwarted the further development of class formation (Cowen and Shenton 1996). In effect, this proposition is an attempt, through the prism of classes of labor, to map Marx's sketch of primitive accumulation on a global scale beyond its effects for the emergence of capitalism in England and other early capitalist economies that he outlined.

In short, to propose that capitalism today as generalized commodity relations is established everywhere is to accept the end of the era of primitive accumulation, even if some of the great variety of forms of capitalist social relations, and the dynamics that underlie them, resemble modalities of that previous era. For example, we know that there are intricate sets of connections between larger- and smaller-scale commodity production in the circuits of many branches of production (see, for example, the study of garment manufacture in India by Mezzadri 2017) and more generally. We also know that “non-economic” relations and practices—social, political, cultural and ideological (not least those [re-]created from “tradition”, “identity”, and so on)—are deployed by both particular capitals and groups among classes of labor to protect or advance themselves. It is misleading to view such common phenomena as either manifestations of the “outside” (or “before”) of capitalism or as indicators of continuing primitive accumulation and of resistance to it, as is well demonstrated in analyses of caste in India by R. Ramakumar (2017) and Alpa Shah and colleagues (2017).

A useful thought experiment in this regard would be to see whether we could take on board the wide range of dynamics and practices held to exemplify (“continuing”) primitive accumulation—ongoing violence and theft in the reproduction of capitalist social relations, “land grabbing”, privatization(s), the ravages of speculative finance capital, the advance of property rights in knowledge, and so on—and to investigate and explain them without recourse to the term at all.

To return to the “population question”, I draw on the recent article by Benanav (2019) which examines aggregated statistical data, much of it generated by demographic studies, to explain “the growth of the global informal workforce, 1950–2000”. In doing so he proposes two forms of “demographic dispossession”: of peasants and of the children of urban workers. The first operates through pressures on access to sufficient land for self-reproduction, as well as class differentiation of small farmers (and “land grabbing” and the like, one should add), evidence of which is provided by many case studies. Concerning the second, he writes the following:

When parents no longer provide for their children, they revisit their own dispossession on their sons and daughters. . . . In periods of population growth, this process issues in an expansion of the urban labor supply, dramatically increasing the size of the dispossessed population. . .[this] was the single largest contributor to proletarianization after 1950, yet it is frequently misconstrued. . . . The key here is that, unlike migration rates. . . rapid population growth rates generate demographic dispossession largely independently of the prevailing speed of economic growth. . .although it unfolded at a decelerating pace, demographic growth continued to give rise to major increases in the labor supply. In the waning decades of the twentieth century, these unfolding trends interacted with a global economic slowdown, issuing in a dramatic informalization of urban work. (Benanav 2019: 686–689)

In short, there is an “autonomy of demographic processes from economic ones”, with “the specific pattern of demographic growth that was behind demographic dispossession” (Benanav 2019: 694, emphasis added) explained in terms of difference from the model of “demographic transition” based on Northern experiences and at the core of macro-population studies (see also Seccombe 1993). In the South “population growth rates were simultaneously much higher and more urban” than in the North (Benanav 2019: 699) as fertility rates did not decline as rapidly as mortality rates, with the latter largely “an unintended consequence of a health transition” (ibid.: 700, emphasis added).

While Benanav's argument is not framed explicitly in Marxist terms, it is informed by materialist concerns, and its dispassionate tone does not exclude recognition of the multiple savage inequalities of global capitalism.⁸ I have chosen to emphasize his “autonomy of demographic processes from economic ones” and “unintended consequence” in support of the position that “surplus population” is better investigated first as one of many contradictory effects of capitalist development, whether in general or manifested in many different circumstances, rather than as “functional” to it by dint of the interests of capital in general or capitalism on a world scale, let alone any conscious intent on the part of the latter. To move from the former to the latter perspective—to investigate whether, how and how much demographic “effects become causes”—requires a great deal more theoretical and empirical analysis than “functionalist” versions of “surplus population” have provided to date.