Workers of the World Unite

The use of Marx to justify nationalist leftism is a well worn genre. This explores the falsehoods this always rests on, including both empirical data on migration and their misreadings of Marx.

There is an entire genre of writing devoted to converting Marx to the nationalist cause. Almost a fortnight after Trump’s inauguration, Monthly Review published an article by David L. Wilson in which he summons Marx in defence of immigration restrictions to the United States (2017). So as to argue that unionists should support Brexit and an end to freedom of movement within the European Union, the British trade unionist Len McCluskey similarly cited a fragment of Marx, in this instance, from a speech to a meeting of the First International (2016). McCluskey and Wilson are not the only ones to have made these arguments in recent years. Some have relied on Wilson’s article in so doing. Angela Nagle drew on Wilson’s claims in “The Left Case Against Open Borders,” for the Catholic nationalist website American Affairs, where she argued for the expanded use of biometric surveillance of migrant workers (2018). Wilson’s article was also republished by Rifondazione Comunista. Parallel arguments for immigration restrictions have been put forward, for instance, by Slavoj Žižek and Wolfgang Streeck (Žižek 2016; Streeck 2018; cf. Mitropoulos 2019).

The claim to leftism that these anti-migrant positions dwell on is that immigration exerts downward pressure on the wages of non-migrant workers. But, firstly, there is no empirical evidence to support that claim. Secondly, there is no evidence that Marx supported restrictions on migration (or nationalism), and a great deal to suggest that he did not.

There is, however, evidence that employers engage in arbitrage. By arbitrage, I mean: employers taking advantage of variations in national currencies, re/production costs and different laws around the world so as to increase profits through such things as offshoring elements of the re/production process or, through immigration policy, by actively creating a class of persons as a fearful, pliant, highly precarious and therefore hyper-exploitable strata of the labour market (Mitropoulos and Kiem 2015).

Arbitrage does not only involve borders. It is also facilitated through other kinds of segmentation, such as the layering of intermittent contracts underneath tenured ones within universities. Use of the term ‘arbitrage’ is more prevalent in discussions of finance, but the mechanism is the same: trading on ranked differences. Contrary to Wilson, but in line with Marx’s analysis and proposed remedies, it certainly does not follow from the existence of arbitrage that it is in the interests of workers—on either side of a border—that migration should be restricted.

It is through a partial account of arbitrage—and some extraordinary elisions of Marx’s analysis and approach—that proponents of immigration restrictions have sought to ‘translate’ a critique of borders into an argument for their preservation and enforcement. Wilson paraphrases long-standing discussions of the importance of stratified legal status made by opponents of immigration restrictions; he also appears to support calls for the legalisation of undocumented migrants; but ultimately concludes that migration should be restricted more effectively, presumably without resorting to xenophobia and “anti-immigrant feeling” along the way. I will address each of these points, though not in that order.

But the crux of it is this: left nationalists deploy a category error at every turn. They replace an analysis of the extraction of surplus value in industries and across the extended circuits of extraction, accumulation and consumption—which involves recourse to arbitrage and has always stretched beyond nation-states—with national aggregates of wage and employment rates that give the impression of a well-defined uniformity among workers within nation-states.

That systematic conflation will become clearer when we look at what is put forward as evidence (and for what), and which of Marx’s words they opt to cite and those which they choose to omit. The most flagrant and consequential omission in the latter regard—which I will come back to in concluding—is the erasure of Marx’s extensive discussion of “relative surplus populations,” which emphasizes “branches of production” (not the “economic fictions” of national measures), and which he developed in the course of his critique of Malthus’ concepts of ‘overpopulation’ and ‘surplus populations.’

By failing to engage with Marx’s writings on this topic, Wilson, McCluskey and others invite us to treat (partial) evidence of arbitrage as if it were evidence that migration places “downward pressure” on the wages of “native-born … workers” as a group.

None of this is meant to imply that Marx’s writings should be used as proof of anything other than Marx’s views and approach. Nor does it suggest that those writings do not fall short—I think that they do, though explaining how they do requires a much longer discussion, including of the history of immigration policy as a defined field of governance which had barely emerged prior to Marx’s death. It is also not a discussion of contemporary neomalthusianism, some of which exists at the intersections of climate change politics, nationalism and white evangelicalism (see Out of the Woods 2017; Mitropoulos 2018; Trafford 2019). It is, instead, an argument that Marx’s analysis of the dynamics through which surplus value is extracted clearly points toward an understanding of borders as mechanisms of capitalist exploitation and, furthermore, that his analysis is irreconcilable with calls to restrict migration.

Marx against nationalism

To make his argument, Wilson cites a passage from a letter, sent by Marx in 1870 to Sigfrid Meyer and August Vogt in New York. While Wilson admits that the letter to Meyer and Vogt may not represent “a comprehensive analysis” of migration, he nevertheless proposes to regard it not only as an insight into Marx’s thinking but his “most extensive treatment of immigration.” This is simply untrue. Marx discussed population movements and their impact on wage and employment levels at great length throughout his writings, including in texts written for publication which appeared during his lifetime.

Wilson cites a passage from the letter which begins:

Owing to the constantly increasing concentration of leaseholds, Ireland constantly sends her own surplus to the English labour market, and thus forces down wages.

Wilson abridges that statement—removing the reference to proletarianisation precipitated by land enclosures—so as to treat it as synonymous with a statement that “the influx of low-paid Irish immigrants to England forced wages down for native-born English workers,” and he does so in order to smooth the inferential leap to nationalism as the solution. He does not wonder whether nationalism is a variety of land enclosure.

Still, there is nothing in the letter to Meyer and Vogt which suggests that Marx’s remedy involved ending migration (either immigration to England or emigration from Ireland). Indeed, in the same letter, Marx contends that nationalism enables “the governments of both countries, whenever they think fit, to break the edge off the social conflict [between workers and employers] by their mutual bullying, and, in case of need, by war between the two countries.” He proceeds to characterise English nationalism as “the secret of the impotence of the English working class, despite its organisation,” that is, “the secret by which the capitalist class maintains its power” by treating the Irish worker “as a competitor who lowers” the “standard of life” of English workers. (Marx [1870] 1975, 221, emphasis in original).

To make his argument for Brexit, McCluskey quotes a fragment from Marx, which runs as follows:

in order to oppose their workers, the employers either bring in workers from abroad or else transfer manufacture to countries where there is a cheap labour force.

What McCluskey left out from Marx’s speech is his clear suggestion as to how workers might respond. In the very next sentence, Marx went on to say:

Given this state of affairs if the working class wishes to continue its struggle with some chance of success, the national organisations [of unions and workers’ associations] must become international (Marx [1867] 2010, 422).

Moreover, neither McCluskey nor Wilson seem to notice that Marx mentions offshoring (as in the fragment cited by McCluskey). That is, even if it were hypothetically possible to end immigration to, say, Britain or the United States (and without thereby increasing the proportion of undocumented workers as a result), it would still be possible to for employers to engage in arbitrage—that is: “to transfer manufacture to countries where there is a cheap labour force.”

Additionally, there is nothing in Marx’s argument that supports the false remedy of ‘restoring’ a geopolitical nationalist order by way of the euphemistic call to “address … the root causes of immigration” (Wilson). Segregation is no more a solution to racism than nationalism is a solution to arbitrage. Furthermore, reframing the event of migration itself as if it were in the interests of capitalists dignifies the power of immigration restrictions and policy while eliminating the capacity of workers to decide for themselves whether to move or stay—the same strategic capacity of workers’ power that is implied in the right to strike, the ability to leave an untenable job, or say no to a contract.

To support this latter set of rights and capacities but not freedom of movement across borders is racist in its discrimination, a self-defeating approach to arbitrage, and does more to enhance the power of governments to control the movements of workers than to increase the power of workers in relation to capitalists, wherever they might be. Marx did not dream of everyone going ‘back to where they came from,’ or argue that globalised nationalism was a solution to capitalist exploitation, as nationalists insist on doing.

In broader terms, Wilson and others also conveniently omit Marx’s critique of economic nationalism and protectionism. Both are clearly stated in numerous writings and, as might be expected of someone leery of abstract equality (or “abstract justice”), are attentive to false equivalences. For instance, in the letter to Meyer and Vogt, Marx supported Ireland’s secession from the British empire and campaigns in defence of Irish prisoners. He did not support English nationalism, or the end of emigration from Ireland. Against Malthusian understandings of ‘surplus populations,’ Marx pointed out that “the Irish labourers left behind” during the period of mass emigration did not experience an improvement in their wages or conditions as a consequence of the decline in population: “the relative surplus-population is today [1867] as great as before 1846,” the “wages are are just as low, … the oppression of the labourers has increased, [and] the misery is forcing the country toward a new crisis,” because the transformation of “agriculture has kept pace with the absolute depopulation” (Marx [1867] 1978, 1:659).

That is, Wilson’s glosses over the significance of “the constantly increasing concentration of leaseholds” in the Meyer-Vogt letter—the processes that creates the category of “low-paid … immigrants.” Doing so invites the implication that “low-paid” is a reflection of some intrinsic value attributable to nationality and birthplace and not, to the contrary, the result of the same processes of proletarianisation and capital accumulation that had already taken place through—as Marx writes in the Meyer-Vogt letter—the “clearing of the agricultural districts of England and Scotland” (Marx [1870] 1975, 221–22).

As for Marx’s views on protectionism, Marx was, on balance, in favour of free trade. His disdain for the “cotton barons and iron heroes” who, “one fine day,” “woke to find themselves transformed into patriots” intent on instituting “protective tariffs, [and] a system of prohibitions of national economy” is more than clear. Marx argued that economic nationalists present a narrowed and spurious choice between “national economy or the rule of private property over nationality” (Marx [1844] 1984, 248).

The evidence, and of what?

First, Wilson outlines some of the research on whether immigration exerts downward pressure on the wages of non-migrant workers, principally those studies undertaken by George Borjas and that of Giovanni Peri, who arrive at vastly different conclusions. He does not point to research by Michael Clemens and Jennifer Hunt which has debunked Borjas’ claims, and which found that the drop in wages Borjas attributed to immigration corresponded to a change in statistical methods rather than a decline in wages (Clemens and Hunt 2019; Clemens 2017). By contrast, the research by Peri and others found that there are no uniform effects across time or on all workers. Instead, they suggest that there are short-term effects within particular occupations and for other migrant workers and, additionally, there are differences over both the short- and long-run. Wilson summarizes Peri: “low-wage immigrants will bring down wages slightly for less educated native-born workers, and will depress wages significantly for other immigrant workers already in the country,” but “over the long run, the average worker’s income goes up by 0.6-0.9 percent for each one percent increase in the immigrant population.”

Secondly, Wilson then goes on to dismiss both Borjas’ and Peri’s research on the grounds that, in their models, “the only factors determining wage levels are labor supply and demand and immigrant workers’ education and skill levels.” To these factors, he adds distinctions in legal status, and points to research by Pedro Orraca-Romano and Erika García Meneses, which shows that Mexican-American workers (irrespective of their legal status or birthplace) tend to suffer discrimination, as evidenced by a pronounced wage gap, in the main through occupational segregation.

Thirdly, from all of this, Wilson somehow concludes that “[i]mmigrant rights advocates may feel it is expedient to cite academic economists like Peri who downplay or deny the downward pressure exerted on wages by the exploitation of undocumented workers,” claims that it is not, and cites Moshe Adler, who argues that many U.S. citizens work in occupations with large numbers of undocumented immigrants and therefore “know firsthand that [exploitation of immigrant workers] puts direct downward pressure on their own wages.”

The issue of legal status is crucial and important. Wilson mysteriously chooses not to engage with analyses of immigration restrictions which have made this exact point, but whose conclusions are at odds with his own. Unlike Wilson, they do not treat evidence of arbitrage as if it were an argument for strengthening the mechanisms that enable arbitrage. Moreover, evidence of discrimination against a group of workers—as shown by Orraca-Romano and García Meneses—is not evidence that the wages of another group of workers decline as a result. Wilson appears to understand that illustrations of arbitrage and discrimination point to a kind of anecdotal or “firsthand … knowledge”, inasmuch as it is not indicative of a generalised condition. Yet he cannot explain why there is a discrepancy between the effects and operations of arbitrage and nationally-averaged wage rates. He instead decides to infer from this that migration “undoubtedly” and “inevitably” places downward pressure on the wages of “native-born … workers” in the same occupations—and leaves it to the reader to make additional inferences consistent with his own. He offers no evidence for this claim, either within an industry or occupation, or ask whether there are parallel shifts of workers who are not, say, of Mexican ancestry (irrespective of citizenship status), to managerial or better-paid positions or other “branches of production” as a result—and may explain some of those longer-term effects identified by Peri.

Instead, he chooses to dismiss opponents of immigration restrictions—and those who do not defer to baseless narratives of the “undoubtedly” deleterious effects of migration on wages—as insufficiently working class and presumably not themselves migrants but, instead, “middle class … advocates.” This implicit identification of working class authenticity with a tolerance for the xenophobic assumptions that support structural racism is why, despite everything else, it is possible for nationalists to cite Wilson approvingly. Moreover, at no point does he ask what the effects of migration are on the wage and employment levels of those who migrate, or what the consequences on wage and employment levels are for those who cannot—as Marx did in his discussion of Irish emigration.

Before moving to a brief discussion of Marx on population, it bears repeating that the distinctive contribution of neo-Malthusianism to population theory is the characterisation of quantitative increases in the size of the population due to non-white migration as if that increase amounted to a qualitative change—that is, purportedly categorically distinct from the so-called natural growth rate of a population, and implicitly presented therefore as a transgression of some ‘natural’ order of national-racial division. Proponents of immigration restrictions who draw on Malthus are not inclined to also demand limitations on white people having children; nor do they tend to support abortion—Malthusianism is, at its core, a reactionary doctrine wrapped in pseudo-scientific pretensions.

Relative v absolute ‘surplus populations’

The disjuncture between, on the one hand, the statistical imagination of national uniformity in wage/employment levels and, on the other, an analysis of how arbitrage functions through transformations of ‘branches of industry’ (elements of the supply-chain) points to the difficulty that Wilson has in offering evidence that “the influx of low-paid … immigrants” results in the depreciation of the wages of “native-born … workers” as a group—and on which cultural-nationalist redescriptions of class depend. The attempt to reconcile the divergence between ‘anecdotal’ and nationally-averaged “knowledge” that so vexes Wilson’s account can, I think, be explained by the categorical difference that separates Marx’s concept of “relative surplus populations” from Malthusian concept of “surplus populations.”

Had Wilson engaged with Marx’s theory (not misread and cherry-picked a passage from private correspondence), he would have encountered a scathing critique of the Malthusian idea that absolute increases in population size are a factor in determining wage or employment levels—that is, why labour supply is not a determinative factor. For instance, Marx did not imply that “the cause [of a decline in wages] was an oversupply of manual labourers,” as Wilson suggests he did. In Capital, Marx discusses, at length, the dynamics of proletarianisation and the mechanisation of a given “branch of production,” of the movements of workers into and out of “branches of production” (not across national borders), and utterly rejects the Malthusian doctrine of ‘surplus populations’ which treats ‘supply’ as the most significant factor.

Marx was clearly at odds with the Protestant cleric’s presentation of population statistics, as if they represented the workings of a divine law: “with this very ‘principle of population,’” he writes, “struck the hour of the Protestant parsons.” Each “historic mode of production” Marx insisted, contra Malthus, involves “its own special laws of population, historically valid within its limits alone” (Marx 1978, 1:591–92). On this point Marx is consistent: the dynamics which give rise to fluctuations in wage and employment levels are independent of variations in the absolute size of a population. He characterises the production of a ‘reserve army’ (unemployment, intermittence, etc) as the creation of “an artificial over-population.” “The decrease of variable in relation to constant capital”—i.e., the decrease in the ratio of wage labourers to machinery—“which goes hand in hand with the development of the productive forces, stimulates the growth of the labouring population, while continually creating an artificial over-population” (Marx [1894] 1959, 3:245). Unlike the absolute number of the population, Marx’s concept of “relative surplus population” is put forward as an endogenous variable in the dynamics of capitalist exploitation. That is, changes in wage and employment levels are unaffected by “the absolute growth [in the size] of the population.” The “relative surplus population” is “surplus with regard to the average needs of the self-expansion of capital.”

Across his writings, variations in wage and employment levels are presented as a consequence of the drive to increase the rate and magnitude of exploitation through changes in the ratio of variable and constant capital, through the struggles around depreciating the price and value of labour power, changes in the the number but not the wages of workers employed in a “branch of industry”, or increasing the proportion of unpaid or below-subsistence labour (1978, 1:581–82).

In the third volume of Capital, he adds a suggestive qualifier. He admits that his schema suffers from the assumption of “competition among labourers and equalization through their continual migration from one sphere of production to another.” Such a “general rate of surplus value” is, he points out, a tendency “assumed by us for the sake of theoretical simplification,” and often “obstructed by practical frictions … such as the settlement laws for farm-labourers in Britain” (Marx [1894] 1959, 3:172). Characterised by seasonal migration, agricultural workers in nineteenth century Britain were increasingly subject to Laws of Settlement that (in accord with Malthus’ prescriptions) restricted their access to parish welfare, throwing the balance of power to landlords, and ensuring a hyper-exploitable segment of agricultural workers. As a nascent immigration policy conducted through welfare restrictions, it thereby enabled the agricultural estates to engage in arbitrage within a given “sphere of production.”

Indeed, Marx describes the conflation between the concepts of “absolute” and “relative surplus populations” as reliant on an “economic fiction.” It is an allusion to national measures of productivity, employment and wage rates, which give rise to the imagination of a national uniformity in wages, productivity, etc. These “economic fictions” lend credence to the idea of an ontological distinction between ‘native-born’ and ‘migrant’ populations. Their “fictional” character explains why every attempt to furnish evidence that migration exerts downward pressure on the wages of “native-born” workers flounders—neither the dynamics of capitalist exploitation nor “branches of production” have ever been confined by national borders. Instead they have used them to their advantage.

As noted, Marx’s writings cannot give a comprehensive analysis of systems that emerged after his death. Nevertheless, his framework clearly opts for an analysis of the extraction of surplus value rather than the “economic fictions” expedient to claims of national representation. In that regard, they give a far more accurate account of variations in wages and employment levels than those predicated on national aggregates.

Extrapolating from his approach, I would argue that contemporary border policies do not give rise to a system of discrete and internally-uniform units called nation-states with wages and conditions shared by all workers within them. Nor do these policies create the ‘space-time compression’ championed by globalisation theory. On the contrary, borders are important mechanisms in the creation of archipelagic systems involving supply-chains that extend across the world, both the “temporal divergences that are indispensable to even the most infinitesimal practices of arbitrage and hedging within financial markets” and the variable geometry (enforced by detainment, deportation and layered contracts) that gives rise to labour arbitrage (Mitropoulos 2015, 173). The reason why migration has been the pronounced exception to ‘globalisation’ and ‘free trade’ is that it pertains to the independent variable within processes of exploitation—a commodity like no other, as Marx would say.

The only effective response to arbitrage, then, involves dismantling the opportunities for employers to engage in it—the approach that Marx pointed toward, but which nationalists either reject, undermine, or subordinate to the imagination of an idealised order made up of well-defined, ranked nation-states managing the global divisions of labour and exploitation on behalf of capital. The ‘movement’ in ‘workers’ movement’ is not a metaphor.


Clemens, Michael. 2017. ‘What the Mariel Boatlift of Cuban Refugees Can Teach Us about the Economics of Immigration: An Explainer and a Revelation’. Center for Global Development, Washington DC.
Clemens, Michael A, and Jennifer Hunt. 2019. ‘The Labor Market Effects of Refugee Waves: Reconciling Conflicting Results’. ILR Review 72 (4): 818–57.
Marx, Karl. 1959. Capital: A Critique of Political Economy. The Process of Capitalist Production as a Whole. Edited by Friedrich Engels. Vol. 3. Moscow: Progress Press.
—. 1975. ‘Marx to Sigfrid Meyer and August Vogt’. In Karl Marx and Friedrich Engels. Selected Correspondence, 220–24. New York: Progress Press.
—. 1978. Capital: A Critique of Political Economy. The Process of Capitalist Production. Vol. 1. Moscow: Progress Press.
—. 1984. ‘A Contribution to the Critique of Hegel’s Philosophy of Right’. In Early Writings, 243–58. London: Penguin.
—. 2010. ‘The General Council of the International Working Men’s Association. On the Lausanne Congresss’. In Marx & Engels: Collected Works, 20:421–24. London: Lawrence and Wishart.
McCluskey, Len. 2016. ‘The Free Movement of Labour Is a Class Question’. In Institute of Employment Rights. London.
Mitropoulos, Angela. 2015. ‘Archipelago of Risk: Uncertainty, Borders and Migration Detention Systems’. New Formations 84 (84–85): 163–83.
—. 2018. ‘Lifeboat Capitalism, Catastrophism, Borders’. Dispatches 1 (November).
—. 2019. ‘Wolfgang Streeck’s Neo-Romantic Sociology’.
Mitropoulos, Angela, and Matthew Kiem. 2015. ‘Cross-Border Operations’. The New Inquiry 18 (November).
Nagle, Angela. 2018. ‘The Left Case against Open Borders’. American Affairs 2 (4).
Out of the Woods. 2017. ‘Lies of the Land: Against and beyond Paul Kingsnorth’s Völkisch Environmentalism’. Libcom, March 2017.’s-völkisch-environmentalism-31032017.
Streeck, Wolfgang. 2018. ‘Between Charity and Justice: Remarks on the Social Construction of Immigration Policy in Rich Democracies’. Culture, Practice & Europeanization 3 (2): 3–22.
Trafford, James. 2019. ‘Against Green Nationalism’. OpenDemocracy, March 2019.
Wilson, David L. 2017. ‘Marx on Immigration: Workers, Wages, and Legal Status’. Monthly Review, February.
Žižek, Slavoj. 2016. Against the Double Blackmail: Refugees, Terror and Other Troubles with the Neighbours. London: Penguin.


Angela Mitropoulos (@Mitropoulos_A)

Angela Mitropoulos is a Sydney-based academic, the author of Contract & Contagion: From Biopolitics to Oikonomia (2012), and has written on precariousness, borders and welfare policy over many years. Her current research project is ‘Infrastructure and Uncommon Forms’.