In Defense of Unequal Exchange: Critique of Paul Cockshott’s First Worldist Ignorance
In Defense of Unequal Exchange: Critique of Paul Cockshott’s First Worldist Ignorance

In Defense of Unequal Exchange: Critique of Paul Cockshott’s First Worldist Ignorance

Print Friendly, PDF & Email

Rob Ashlar offers a rebuttal to Paul Cockshott’s attack on the theory of unequal exchange. 

Cover of A. Emmanuel’s ‘Unequal Exchange’ (Siglo XXI de España Editores, Madrid, 1972)

Wherever wages are high, universally throughout the whole world, it is an infallible evidence of the riches of that country; and wherever wages for labour run low, it is proof of the poverty of that place.

 

–– Josiah Child, 1668

Cockshott and Unequal Exchange

A few years ago, Paul Cockshott published a series of videos in which he attempted to refute unequal exchange.1 His failure to do so is representative of most self-styled Marxist replies to Arghiri Emmanuel,2 who theorized the concept, so it is worth discussing in some detail. Cockshott begins by asserting that unequal exchange as a theory is both anti-Marxist and anti-worker, so it must be rejected. From this conclusion, he works backwards to build his argument, which he bases on comparisons of productivity between Indian and American laborers, his respective stand-ins for Third World and First World labor. These comparisons reveal as much about Cockshott’s tenuous grasp on Marxism as they do about the structure of unequal exchange.

Cockshott claims that Indian workers earn much lower wages than American workers because they are less productive than the latter–the classic anti-unequal exchange argument. He takes two industries, steel and agriculture, as case-studies, but we will focus on steel since the comparison is simpler than that of agriculture.3 By calculating steel tons produced per capita, he finds that American workers are 6.7 times more productive than Indian workers. Although the calculation is crude, the conclusion is likely accurate.4 The American steel industry has several major advantages, most important of which are skilled labor (high organic composition of labor, OCL) and advanced machinery (high organic composition of capital, OCC). Together, these factors would make the average American steel worker more productive than his Indian counterpart, justifying a wage differential which, at most, corresponds to the productivity differential. Does this correspondence obtain in reality? Cockshott implies that it does, but he omits the wage rates which would show such a correspondence.

Reports by the US Bureau of Labor Statistics reveal that American workers are paid twenty to forty times more than Indian workers, far exceeding the productivity differential.5 Let us examine manufacturing in general before focusing on the steel industry. Indian manufacturing has two sectors, organized and unorganized, which respectively employ 20% and 80% of the industrial workforce.6 ‘Organized’ refers to activity which is formally registered with the government, thus subject to monitoring and regulation. This sector produces over two-thirds of Indian manufacturing output and is most comparable to the industrial sectors of advanced capitalist nations, so we will narrow our view to it. The Bureau found that in 2005, the average hourly wage for all manufacturing employees was $0.91–or 3.1% of corresponding American wages–and for just production workers, was $0.63–or 2.6%.7 By 2010, these figures had increased to $1.46 and $0.92, respectively.8 For 2005, the respective differentials would roughly be factors of 32 and 38.5, and for 2010, they would be 20 and 32, assuming no growth in American wages. We obtain similar figures for the same years when we restrict our calculations to the Primary Metals (PM) and Fabricated Metal Products (FMP) sectors, which together encompass steel.9 As of January 2023, American hourly wages in these fields were $32.38 and $29.14.10 The International Labour Organization reports that since 2008, Indian wages have grown by slightly less than 40%.11 Suppose that wage growth in Indian manufacturing, particularly in steel, followed this trend. For ease of calculation, also suppose that growth was a neat 40% since we take 2005 as our starting year for Indian wages in the PM and FMP industries. Under these assumptions, we find that hourly wages have respectively grown to $1.75 and $1.50, yielding differentials of 18.5 and 19.4.12 Recall that this pertains to a small part of the organized manufacturing sector, which itself carries only 20% of industrial employment. The unorganized and agrarian sectors–that is, highly cheapened workforces–exert downward pressure on wages. Furthermore, although Indian steel is making advances, the country faces steady deindustrialization as part of a global trend in the Third World. This, too, acts negatively on wages.13 In short, our estimate for present differentials is likely optimistic. In every year and every scope, the wage differential exceeds any productivity differential by many factors. The former must have a cause other than the latter. 

If the wage differential between American and Indian workers owes little to productivity, the cause must be something else, namely, politics. Let us emphasize just how anomalous this political-economic conjuncture is. For most of history, wages were–and for most of mankind, remain–permanently fixed to subsistence. Lasalle’s ‘iron law of wages’ had motioned towards a real historical phenomenon, but at the time, one increasingly giving way to the ‘historical and moral element’ of wages. In Arghiri Emmanuel’s words:

Something happened towards the end of the nineteenth century and this conventional Marxist picture [of wages fixed at subsistence cost] ceased to reflect reality. This was a radical change in the distribution of power among the classes within the bourgeois parliamentary system in the industrialized countries, which had the effect of finally lifting the price of manpower out of the swamp of the mere physiological survival of the worker. In order to appreciate the effect of this event it suffices to remember that, for thousands of years, the worker’s wage was the most stable economic magnitude which existed. The purchasing power of the average European worker around about 1830 differed very little from that of the worker in Byzantium, Rome or the Egypt of the Pharaohs. In the following century and a half it was multiplied by ten [emphasis mine].14

This process was intensified in the white settler-colonies, where, from the start, workers’ wages exceeded subsistence–Werner Sombart’s study of the US shows this in detail.15 Indeed, these societies best show that the link between productivity and wages is a tenuous one. Compared to their British counterparts, the white invaders of the early US, Canada, or Australia were not very productive. While the English worked in the nascent mills and factories, the white settlers worked with household handicrafts, yet it was the latter whose wages were several times more than the former’s. The access to abundant land stolen from the native peoples released pressure on the labor market. This forced up the price of labor, which persisted long after said access had closed. Attempts to decrease this price were met with bitter but successful resistance. For example, when looking at Australia, it is difficult to ignore how this country whose industry is quite simple–mere mineral extraction and livestock rearing–has always paid its workers better than almost anywhere else on Earth. Conversely, when looking at England from 1780 to 1850, it is just as difficult to ignore how this country with ever-advancing industry–the ‘workshop of the world’–paid its workers ever-decreasing real wages. Many good historical studies have shown that during England’s industrial revolution, workers’ living standards considerably fell.16 It was not until workers began to fight in earnest, by forming trade unions and the Labour Party, that they could take a larger share of the social product. Consider also that in response to the First and subsequent Congresses of the Communist International, Western elites granted workers’ the proto-welfare state.17 Class struggle is decisive in improving workers’ lot.18 For:

Wage labour must be regarded as it is: a social relation in which classes fight for their interests–i.e., the struggle for the division of the social product into wages to the labourers and profit to the capitalists. Thus wages constitute a part of the social product and their size reflects the relative strength between the classes and the economic basis on which the class struggle takes place.19

Herein is the ‘political’ in ‘political economy’. In contrast, Cockshott’s argument implies that it is always in the interest of workers for their bosses to invest in new machinery, to which the corollary is that their bosses’ and their own prosperity is one and the same. Such revolutionary ‘Marxism.’

Cockshott not only equates wages with productivity, but even claims (7:40 in ‘The so-called unequal exchange’) that the higher productivity of American steel workers makes them more exploited than Indian workers. His argument goes like this: on the global market, the steel products of two countries obtain the same price. The more mechanized country’s steel will have a smaller amount of labor, so its workers create proportionately more value, making them more exploited.20 This argument rests on two flawed assumptions. First, it presumes that workers own the machinery necessary for their productivity, when this belongs to their bosses. To claim that productivity increases directly translate to wage increases, suggests that the profits derived from the latter are already owned by workers. In fact, productivity gains are the result of productive investments made by capitalists, financed by profits, which by definition belong to the bourgeoisie. Thus, Engels wrote to Lafargue:

…in what respect the wage worker gains an advantage in seeing his productivity increase, when the product of that productivity does not belong to him and when his wage is not determined by the productivity of the machine [emphasis mine].21

In short, Cockshott’s argument is irrelevant to basic capitalist reality. Second, it presumes that workers in both countries are paid equally. Under that condition, it holds true, for it describes the broader effect of the development of the productive forces. All else held equal, this applies downward pressure on wages.22 However, note that Cockshott’s invoking of this argument contradicts his own earlier claim that American and Indian wages reflect their productivity–that is to say, all else not held equal–which implies that the ratios of labor cost to exchange value are the same.23 For our part, we have shown that both cases–equal wages and productivity-linked wages–are false, so the argument is moot. American wages are in a league of their own, which, as we will see, would be impossible for American capitalism to tolerate were it not for imperialism. 

Wages and Development

We must now show the relationship between high wages in one country and low wages in another–that is, unequal exchange. Cockshott denounces the ‘notion that low wages are explained by unequal exchange’ (10:10 in ‘The so-called unequal exchange’), suggesting a weak grasp of the subject. It is not unequal exchange which leads to low wages, but the other way around. As Amal Samaha put it:

If there are two or more trading partners that are competing for the same mobile pool of capital [i.e. under conditions of globally equalized rate of profit], a general wage rise in one country without a corresponding increase in the surplus value contributed by labor to its commodities will result in that country having a higher price of production (which we must remember does not necessarily correspond to unit prices, but to the sum of prices for a given quantity of invested capital) for its total production relative to total commodity values, regardless of the initial value composition of its industry. In contrast, all other countries which did not see such a rise in wages will see lower prices of production relative to initial commodity values. […] The capitalist of the richer country gets access to cheap elements of constant capital from the poorer country, while the poorer country must pay even more for high-value-added goods from the core.24

This is a dynamic, historical model with several implications. First, it is not certain commodity types as such, but commodities made by expensive labor, which command high prices. Second, low- and high-wage commodities, by extension industries, must be inter-sectoral. Once a low-wage country effectively competes in a sector dominated by a high-wage country, the latter switches to another. The original sector’s prices fall, the new sector’s prices rise, and the structure is reproduced anew. Third, the high wages in one country directly depend on the low wages in another. Returning to the example of steel, under ‘normal’ conditions, American workers’ high wages would rapidly cut into the rate of profit, disturbing global equilibrium and forcing a mass exit of capital from the industry. Therefore, steel corporations compensate by importing cheapened production inputs such as Brazilian iron ore, which in turn forces the Great Lakes iron ore industry to cut costs by increasing labor productivity through capital investments and new methods.25 Fourth, unequal exchange reinforces but does not cause unequal development, for both have the same source: unequal wages. Cockshott unwittingly concedes this point when he notes that low Indian wages discourages productive investments and upskilling. Low wages (re)produce low OCC and OCL, and conversely, high wages (re)produce high OCC and OCL–the cycle infinitely repeats. Further, since the individual capitalist’s aim is not to produce but to sell, he goes where there are ample markets created by high disposable income. It is unsurprising, then, that most foreign direct investment goes to rich countries, where there are large markets, skilled workforces, and stable societies. Indeed, most capital built up in the third world is eventually invested in the first world by the same mechanism, according to the same logic.26 High wages therefore hoover up capital, over-developing some nations by under-developing others. Put another way, in the capitalist system, wages are the final determinant of development27 – that is, whether some nations will starve while some fatten. As Emmanuel wrote: ‘What is involved here is something quite irrational and backward-looking, namely, hereditary privilege.’28

Opponents and Supporters

It must be noted that the exchange above is merely one repetition of the same debate which began with the publication of Unequal Exchange over fifty years ago and has continued since. With his videos, Cockshott has reduced the anti-unequal exchange position to its barest elements, revealing just how fragile and even stupid those elements are. He adds nothing to the discussion and joins the long list of Emmanuel’s weak opponents, some of whom we will briefly review now. 

As mentioned in my second footnote, the first and most comprehensive ‘Marxist’ reply was by Charles Bettelheim, whose view is stated in Appendix I of Unequal Exchange. His argument (the full content of which we cannot repeat here) is a more sophisticated-sounding version of Cockshott’s. His explanation for wage differentials and global development patterns are each country’s ‘specific combination of productive forces and productive relations,’ which is the Althusserian term for productivity. The body and subsequent Appendices of Unequal Exchange adequately debunk this and Bettelheim’s other claims. Cockshott’s debt to Bettelheim goes without saying.

The next major Marxist reply to Emmanuel came from Samir Amin, most comprehensively stated in ‘The End of a Debate.’29 Despite common misconceptions, Amin and Emmanuel strongly disagreed on political economy.30 The former opposed the latter’s theory of unequal exchange, with which he had several fundamental disagreements, particularly his theory of wages and of international exploitation. He (arrogantly) writes: ‘My analysis of the transfer of value is superior to that of Emmanuel; it is the only analysis that permits a correct definition of unequal exchange: the exchange of products whose production involves wage differentials greater than those of productivity.’31 Despite clever turns of phrase, Amin’s achievement was not to produce a more precise analysis of unequal exchange–his version is unrigorous and incoherent–but to shift the blame, as it were, for unequal exchange from workers in rich countries to multinational corporations. This makes the theory compatible with and palatable to mainstream Marxist analysis of imperialism, which could explain Amin’s greater fame and staying power.  

At around the same time as Amin, the famous bourgeois economist Paul Samuelson wrote his own reply to Emmanuel, called ‘Illogic of Neo-Marxian Doctrine of Unequal Exchange.’ Like almost all critics, Samuelson did not understand Emmanuel’s theory, owing to his apparent failure to read it. His argument accordingly suffers from confusion. He attempts to debunk Emmanuel under a different set of assumptions, making the conclusions derived thereof irrelevant. Thus, Samuelson’s achievement is to be the most prominent opponent to Emmanuel who has not understood, perhaps even read, the theory in question–Cockshott has also taken after this example.

The confusion is not just with opponents but even supporters of the theory, such as Zak Cope and Jason Hickel. Both make the same mistake, namely, to attempt calculations of ‘value transfers’ arising from unequal exchange. This is perhaps where Cockshott gets the misconception that his opponents ‘explain low wages with unequal exchange,’ as though these exchanges directly lead to high or low wages. Cope makes this mistake in his books, Divided World Divided Class and The Wealth of (Some) Nations, while Hickel does so in various research papers written over the past few years.32

For his part, Emmanuel was not concerned with such calculations and used them only for illustration. The value of his theory was its use as a dynamic, historical model to explain past and current global development trends. Writers like Cope and Hickel have arguably confused more than they have enlightened, despite good intentions.33 This can be seen in Cockshott’s second and third videos (‘Again on Indian productivity’ and ‘Mr Mittal’, respectively) in which he responds to a dissenting commenter, influenced by Cope, who clumsily argues that American steel workers are considered more productive because they are using recycled scrap in which the ‘value-added’ is from years or decades prior. Such sloppy arguments only harm the anti-imperialist position. However, since Cockshott has portrayed himself as a rigorous Marxist thinker, one cannot justify his failure to go beyond vulgarized, secondary interpretations of Emmanuel’s theory. His error reveals that not only does he not understand unequal exchange–owing to an apparent inability to read his opponents–he also does not understand basic Marxism (e.g. productivity). His videos highlight the importance of returning to Emmanuel’s particular theory of unequal exchange due to its rigor and utility in contrast to its vulgar popularizations and its even more vulgar First-Worldist opponents.

 

 

Liked it? Take a second to support Cosmonaut on Patreon! At Cosmonaut Magazine we strive to create a culture of open debate and discussion. Please write to us at CosmonautMagazine@gmail.com if you have any criticism or commentary you would like to have published in our letters section.
Become a patron at Patreon!
  1. Video 1, Video 2, Video 3. Note that Cockshott is a rank transphobe, so my response is already more generous than necessary.
  2. Including the first reply, written by Charles Bettelheim, who advanced the same argument as Cockshott’s but in a more sophisticated form than his. It is reproduced as Appendix I in: Arghiri Emmanuel, Unequal Exchange: A Study of the Imperialism of Trade (New York: Monthly Review Press, 1972).
  3. In agriculture, we would have to control for the widely differing class statuses of farmers in India and the US. However, it should be noted that were Cockshott’s argument correct (we will show that it is not), it would imply that Argentinian farmers make many times more than any other since Argentinian farmland is perhaps the most productive in the world. Actual reality states otherwise, suggesting something other than productivity is at work.
  4. However, note that there are two problems in this line of argument. First, it implies workers own their productivity, which actually belongs to the capitalists, by virtue of private property (we will discuss this further below). Second, even if workers owned their productivity, this view can only apply to manufacturing labor, where productivity differentials can be empirically measured. It cannot apply to service labor where productivity gains are minimal and differentials highly nebulous. It is difficult to justify why an American cashier should make $30,000 per year while an Indian cashier makes $1,000-2,000 per year when they both perform the same work. Wage differentials narrow as one looks at higher skilled sectors, e.g. tech, but are still quite large, and in some cases are perplexing. For instance, in major professional services companies, entry-level (‘analyst’) positions in the US are paid as much as management positions in India, even though the former is clearly less important to business operations than the latter. There is no justification beyond national privilege.
  5. However, many Indian steel works are approaching Western quality and productivity standards, narrowing the differential, thus eliminating it as a justification for wage disparity. Further, there are many industries in which Third World and First World labor are equally productive but unequally paid. In general, manufacturing labor productivity undergoes what economist Dani Rodrik calls ‘unconditional convergence’–that is, equalization of productivity in both rich and poor countries (cf. Rodrik, “Unconditional Convergence in Manufacturing,” Quarterly Journal of Economics 128, no. 1 (2013)). Cockshott’s ‘productivity argument’, were it even valid, would be irrelevant in this context.
  6. Unless otherwise stated, all information about Indian manufacturing is extracted from: Jessica R. Sincavage, Carl Haub, O.P. Sharma, ‘Labor costs in India’s organized manufacturing sector’, Monthly Labor Review (May 2010). Link.
  7. Sincavage et al take $29.74 as the average American manufacturing wage in 2005.
  8. ‘India’s Organized Manufacturing Sector’, US Bureau of Labor Statistics. Accessed: 27 Feb. 2023, Link.
  9. This is implied by Chart 6 in ‘Labor costs…’ and Chart 3 in ‘India’s Organized…’.
  10. ‘Primary Metal Manufacturing: NAICS 331’, Industries at a Glance, US Bureau of Labor Statistics. Accessed: 27 Feb. 2023, Link; ‘Fabricated Metal Product Manufacturing: NAICS 332’, idem. Accessed: 27 Feb., 2023, Link.
  11. Chart 3.5 in: Global Wage Report 2022–23: The impact of inflation and COVID-19 on wages and purchasing power (Geneva: International Labour Office, 2022), 52. Link.
  12. The hourly wages were determined as follows then rounded to the hundredth decimal: 29.74 * 0.042 * 1.4 = 1.749 and 29.74 * 0.036 * 1.4 = 1.498. In general terms, we may write: (American wage in 2023) * (Indian wage as a percentage of American wage in 2005) * (growth of Indian wage since 2005) = (Indian wage in 2023). I ask the reader to forgive my crude work since this is merely for illustration. The lack of precise data means I have to over-simplify.
  13. Cf. David Oks and Henry Williams, ‘The Long, Slow Death of Global Development’, American Affairs 6, no. 4 (Winter 2022). Link.
  14. Arghiri Emmanuel, ‘The Multinational corporations and inequality of development’ UNESCO International Social Science Journal 28, no. 4 (1976): 761. Link.
  15. Cf. Werner Sombart, Why Is There No Socialism in America? (London: Macmillan Press, 1976).
  16. For one study, see: Eric Hobsbawm, Labouring Men (Garden City: Anchor Books, 1967), 75-149.
  17. Cf. Magnus B. Rasmussen and Carl H. Knutsen, Reforming to Survive: The Bolshevik Origins of Social Policies, (Cambridge: Cambridge University Press, 2022). For the appendices, see the following: Link.
  18. Note that this class struggle need not be progressive, as shown in settler colonial societies.
  19. Communist Working Group (henceforth, CWG), Unequal Exchange and the Prospect of Socialism (Copenhagen: Manifest Press, 1986), 70.
  20. This claim also goes back to Charles Bettelheim. A full rebuttal can be found in: CWG, op. cit., 71-77, 116-122.
  21. Quoted in: CWG, op. cit., 74.
  22. Marx describes the social effect of this pressure in these terms: ‘The greater the social wealth, the functioning capital, the extent and energy of its growth, and therefore also the greater the absolute mass of the proletariat and the productivity of its labour, the greater is the industrial reserve army. […] The more extensive, finally, the pauperized sections of the working class and the industrial reserve army, the greater is official pauperism. This is the absolute general law of capitalist accumulation [emphasis his].’ Arghiri Emmanuel adds: ‘History has proceeded, too, as if the industrialized [i.e. rich] countries had succeeded in exporting impoverishment so effectively that the forecasts of Marxism, which have begun to show signs of losing reality within the context of the industrial nations, are being realized to perfection on the scale of the world economy [emphasis his]’, Unequal Exchange, 263.
  23. Examine: Suppose that wages directly correspond to productivity. This implies that if one steel worker is 6.7 times more productive than another steel worker, he is 6.7 times more costly to the bosses. His higher wages are compensated by the higher value he creates, thus preserving the original ratio of labor cost to exchange-value. It is possible that the first worker’s wages are 5 times greater than the second’s. Technically, he would be more exploited while being more productive than the second, in which case the ratio has changed (this would still be irrelevant due to actual differentials). However, Cockshott does not make this case, so we will not grant it to him.
  24. Amal Samaha, ‘The Mature Labor Aristocracy and Its Problems, Part 2’, Cosmonaut Magazine. Accessed: 2 March, 2023, Link. She explains Arghiri Emmanuel’s own rendition of the theorem, which reads: ‘If the wage is exogenous (institutional, independent variable), and if a tendency exists for the formation of a general international rate of profit, then any autonomous variation in the wage-rate in one branch or in one country will entail a variation in the same direction of the respective price of production and a variation in the opposite direction of the general rate of profit.’ See: ‘Unequal Exchange Revisited’, IDS Discussion Paper, no. 77, (Brighton: University of Sussex, 1975), 34. Link.
  25. Cf. James A. Schmitz Jr., ‘What Determines Productivity? Lessons from the Dramatic Recovery of the U.S. and Canadian Iron Ore Industries Following Their Early 1980s Crisis’, Journal of Political Economy 113, no. 3 (June 2005). Link.
  26. Here is one estimate of the scale of capital flight from the Third World: ‘net financial outflows across all low- and middle-income countries amount to approximately $2 trillion annually, with over $16 trillion having been drained out in total since 1980’ (Elias Nosrati et al, ‘Structural Adjustment and the Political Economy of Capital Flight’, Socio-Economic Review). For a broader analysis of capital flight, see: Gerald A. Epstein, ed., Capital Flight and Capital Controls in Developing Countries (Northampton: Edward Elgar Publishing, 2005).
  27. The Swiss bank UBS agrees, recently reporting that ‘if we are to avoid a 1930s downturn, real wages [in the US] have to turn positive [emphasis his].’ In general, the bulletin confirms Emmanuel’s theory of wages (cf. Paul Donovan, ‘Surveying the weird’, UBS Insights, 7 July 2023. Accessed: 8 July 2023, Link).
  28. Unequal Exchange, 426.
  29. This is Part IV of Imperialism and Unequal Development (New York: Monthly Review Press, 1977).
  30. For an account of their disagreements, see: John Brolin, Bias of the World: A History of Theories of Unequal Exchange from Mercantilism to Ecology, PhD diss. [draft version], (Lund University: 2006), 296-309. Link. For a broader critique of Amin’s theory of political economy, see: Sheila Smith, ‘The Ideas of Samir Amin: Theory or Tautology?’, The Journal of Development Studies 17, no. 1 (1980). Link.
  31. Amin, op. cit., 211.
  32. Cf. Hickel et al, ‘Imperialist appropriation in the world economy: Drain from the global South through unequal exchange, 1990–2015’, Global Environmental Change 73 (March 2022), Link; and Hickel et al ‘Plunder in the Post-Colonial Era: Quantifying Drain from the Global South Through Unequal Exchange, 1960–2018’, New Political Economy 26, no. 5 (2021), Link.
  33. That is, at least from Hickel, whose sincerity is clear. Cope seems more nefarious. As Amal Samaha put it: ‘It’s true that some authors who discuss UE [unequal exchange] are of an anti-Marxist left-liberal persuasion, attracted to a particular interpretation of UE so as to hide a do-nothing program of small social reforms behind ultraleft rhetoric. Cope’s recent article in his Oxford Handbook of Economic Imperialism is a good example of this politics: Zak Cope, “Imperialism and Its Critics: A Brief Conspectus” in The Oxford Handbook of Economic Imperialism, ed. Zak Cope and Immanuel Ness (Oxford: Oxford University Press, 2022), pp. 15–42.