Effective Altruists in praise of Sweatshops

An emerging new way of doing good is the one proposed by the advocates of Effective Altruism, the philosophical and social movement founded by Peter Singer and William MacAskill. As the name Effective Altruism suggests it is constituted of “two parts”[1]: altruism, by which is simply meant “improving the lives of others”[2] and effective by which is meant that the good that should be done it is not just “some good” or whatever amount of good, but “the most good with whatever resources you have”[3]. Typically, an effective altruist (or “aspiring effective altruist”)[4] is a person who donates to charity organisations in the most effective possible way in order to achieve the highest possible amount of good. Needless to say, whether and how it is possible to measure the effectiveness of a course of action or the success and effectiveness of a charity organisation is a matter of debate. Because of its consequence-based approach, EA is particularly compatible with a consequentialist ethical framework[5]. The movement sprung out of philosophy and relies on philosophical research in order, for example, to assess which causes and interventions should be prioritised and being able to effectively address pressing social issues of our time such as global poverty, famine, climate change, animal welfare and existential risk.

I will make an example to prove that Effective Altruism is not neutral, but it usually embraces liberalistic views on certain matters (such as sweatshops); I will also clarify that liberalistic views are usually unable to see the role played by different forms of social action (such as unionisations), but also institutions emanating social legislation and collective engagement in general; showing that EA, in a similar vein, is unable to see the contribution non-private actors and other political or social forces might give in doing good.

In Chapter 8 of his book Doing Good Better MacAskill examines the possibilities of individuals as private consumers in relation to sweatshops, Fairtrade, low-carbon living and vegetarianism[6]. What MacAskill believes every individual can actually do is:

– using her purchasing power to buy (or to boycott) products manufactured in sweatshops;

– donating (or not donating) to a charity recommended by EA;

– adopting a more environmental-friendly life-style which might include stopping buying animal products;

I will leave aside his discussion about vegetarianism and animal rights as well as his criticism against Fairtrade and I will focus on MacAskill’s argument in favour of sweatshops, which I think is the most relevant topic in Chapter 8 for showing what is wrong with Effective Altruism:


Sweatshops are factories in poor countries, typically in Asia or South America, that produce goods like textiles, toys, or electronics for rich countries under pretty horrific working conditions. Workers often face sixteen-hour-days, six or seven days a week. Sometimes they’re prohibited from taking meal or toilet breaks. Air conditioning is rare, so factories can be very hot. Health and safety considerations are commonly neglected and employers sometimes abuse their workers. Because conditions in sweatshops are so bad, many people have pledged to boycott goods produced in them […]. This movement has noble intentions: the people who campaign against sweatshops are justifiably horrified that people work in such awful conditions. However, those who attempt to combat sweatshops by refusing to buy goods produced in them are making the mistake of ignoring the key question […]: What would have happened otherwise? […] In developing countries, sweatshop jobs are the good jobs, the alternative are typically worse.[7]


It follows that boycotting products from sweatshops would actually harm those people whom we are supposed to help.

To corroborate his point MacAskill quotes the interview of a Cambodian woman who scavenges plastic from dumps and says that she would love to get a job in a sweatshop factory. Yet, MacAskill asserts that those jobs are free and consensual exchanges between employers and workers (“almost all workers in sweatshops choose to work there, and some go to great lengths to do so”)[8]. In MacAskill’s view the formal expression of consent is enough to make an exchange free. But he is making the same mistake as some liberals whose commitment to a subjective theory of value carries the risk of considering every transaction where consent has been giving and no one else has been harmed as free and devoid of wrongdoing[9].

A short digression might be necessary here in order to remind the readers of what effective altruists and liberals have in common. Effective altruists are arguably obsessed with “neutrality” and doing good through rationalism and calculations. Timothy Syme, one of the best critics of Effective Altruism poses the following question: how would an effective altruist assess whether the entire capitalistic system is actually based on free exchange or structural coercion?[10] Apart from the fact that it might be unlikely for an effective altruist to ask this question, the answer she would give might be that the best way to evaluate the whole capitalism is to count the number of free and coerced labour contracts. However, as Syme points out, this method is fallacious. In order to evaluate the rightness or wrongness of a socio-economic relationship, in fact, we cannot just make use of “neutral” calculations: it is necessary to first specify what is freedom and what is coercion, which is “an interpretative task, requiring historical and normative inquiry into the character of wage labour”[11]. A similar argument might be made when evaluating democracy: the fact that all citizens have right to vote does not necessarily imply that they will use their right in the best way nor that they cannot be manipulated by political propaganda. Thus, assessing the quality of democracy of a country by exclusively counting the number of electors exercising the right to vote is a mistake. Likewise, with regards to the moral problem of determining if a sexual act is actually consensual we cannot simply ask the participants if they agree without problematizing the socio-economic power relationships at play nor taking into consideration the specific context where the act takes place: the complexity of the notion of consensus cannot be left aside. Determining whether capitalism as a system of socio-economic relationships is just or unjust and evaluating the quality of democracy just as judging the consent of an action, is an interpretative task. In the light of these attitudes, the effective altruist apology of sweatshops comes as no surprise.

Yet, on the basis of MacAskill’s assertion that “in developing countries, sweatshop jobs are the good jobs, the alternative are typically worse[12], it is not clear why we should conclude that the workers freely choose those jobs, rather than concluding that it is more plausible that the workers are forced to work in sweatshops because this is the least bad option. But the fact that working in sweatshops is better than working in the agricultural and informal sectors is not enough to deny their oppressive and exploitative conditions. The mainstream economic argument of free exchange between two free individuals (the employer and the employed) is irrelevant for determining if a factory is sweatshop. The nature of the exploitation is, in fact, determined by the effective characteristics of a job[13]. Remarkably enough, at p. 163 of his book MacAskill states that, while we should not boycott sweatshops, the correct response is to “try to end the extreme poverty that makes sweatshops desirable places to work in the first place”[14]. However, this statement is vague and unqualified. Once again, MacAskill’s proposal is focused on what the individual alone can do: he suggests to buy products from companies that claim to have higher labour standards (such as People Tree, Indigenous and Kuyichi)[15], but he does not say what he actually means by “higher labour standard”, how do we actually know if they are respected and, more importantly, how to guarantee the enforcement of higher labour standards to other companies through national or international legislation. MacAskill’s proposal is insufficient for the same reasons why it is insufficient (at best) and counterproductive (at worst), merely donating to effective charities without considering the possibilities of social and institutional changes and by taking into account solely low-hanging-fruits. On the contrary, the relevant question for a movement aiming to actually doing the best good should be how to promote social, political and economic changes able to empower those workers and free them from the coercion of an exploitative job. These changes can be achieved by neither consumer pressure nor private donations to effective charities. If successful, they might imply the complete eradication of sweatshops. But they are incompatible with at least two of EA’s current beliefs and practises: the use of the strictly economic law of marginal returns[16] to assess “effectiveness” and the belief that capitalism is good and that socialism is incompatible with human nature[17]. That is why MacAskill cannot be credible when he writes that “the correct response is to try to end the extreme poverty that makes sweatshops desirable places to work in the first place”[18].

Although MacAskill’s and many other effective altruists, at least in some cases, arguably seem to embrace a subjective theory of value (compatibly with the belief EIC3), sweatshops could actually be defended on the mere basis of their economic utility. As argued by Ferguson, “whether a particular transaction is exploitative is, in principle, a separate question from whether it ought to be prohibited from either a moral or legal perspective”[19]. On this ground, it might be acknowledged that sweatshops are exploitative while at the same they might be defended on the basis of their economic function. However, the passage about the Cambodian woman[20] shows that MacAskill believes that that transaction is not exploitative. MacAskill, in fact, seems to agree with Zwolinski that “although the employers are treating their employees in some respects as means by benefitting from their labour, […] such treatment is ubiquitous and generally untroubling [since] failure of respect occur when people are used without their free consent”[21]. Yet, even if MacAskill admitted that the mainstream neoclassical standpoint of free exchange contract is mistaken, he would still have a very appealing argument on his side: the purely economic utilitarian argument shared, among others, by the Nobel-laureate Paul Krugman:

The reason there’s such widespread support among economists for sweatshops is that low wage, labour-intensive manufacturing is a stepping stone that helps an economy based around cash crops develop into an industrialised, richer society. During the Industrial Revolution, for example, Europe and America spent over a hundred years using sweatshop labour, emerging with much higher living standards as a result.[22]

This passage, perhaps more than anything else in MacAskill’s book, exemplifies the problem EA has with the interpretation of broad social phenomena. Once again, with regards to interpretative tasks, EA is impartial in theory, but in practise it either lacks, wherever it is necessary, a social analysis and it avoids dealing with broader phenomena or, whenever it takes a view on something, it is the view of mainstream economy, namely a liberalistic view[23]. According to EA-liberalistic view, the market forces alone will be able to adjust the evils of capitalism. Thus, sweatshops are stepping stones less developed economies have to pass through. However, according to the Nobel Prize‑winning economist Simon Kuznets the story of market‑led social progress is questionable:

In democratic societies the growing political power of the urban lower income groups led to a variety of protective and supporting legislation, much of it aimed to counteract the worst effects of rapid industrialization and urbanization and to support the claims of the broad masses for more adequate shares of the growing income of the country.[24]

Kuznets maintains that social progress is the result of the combined effect of economic growth and social legisla­tion. In turns, social legislation is, at least in part, a result of social action or “enlightened collective choices that defenders of sweatshops find objectionable”[25]. Mills argues that the economic and social history of United States makes that clear. For instance, The Fair Labour Standards Act, a set of norms imposed in 1938 by the US government on the garment industry and other industries, was the result of the unionisation that followed the tragedy of the Triangle Shirtwaist in New York. Since then the sweatshops in US had been significantly marginalised, but their number increased again during the 1980s when the wage-and-hour inspectors were reduced by the Reagan administration from 1600 to 700[26]. According to Mill:

The resurgence of sweatshops in the United States [in the 1980s] underlines the importance of political rules and enforcement and makes clear that economic development may increase overall income levels, but it will not by itself eliminate inhuman working conditions. In addition, the return of sweatshops casts doubts on the depiction of sweatshops as a stage that countries pass through just once, as their markets expand and productivity increases.[27]


Thus, Miller maintains that the liberalistic-view that economic development alone brings social improvement “distorts the historical record and misrepresent how social improvement is brought about with economic development”[28]. The liberalistic view might be considered a correlation fallacy (“a causal-logical fallacy where two events occurring together [economic development and improvements of the life-standards] are taken to have established a cause-effect relationship”[29]). On the other hand, it might be argued that asserting that social legislation or class struggle (A) cause social development or improved life standards (B) is equally far from being unproblematic. Thus, the relation between A and B might be also considered a correlation fallacy. It is undeniable that, as long as the problem of establishing historical causation[30] remains largely unsolved, both the “liberalistic” and the “socialist” standpoints will suffer from an epistemological deficit. However, it might be replied that the deficit of the liberalistic standpoint is more serious in so far as many strict liberals completely ignore the contribution of class struggle, social legislation and other social factors and completely neglect alternative social analysis. On the opposite, socialist or anti-capitalist analysis do not usually deny the achievements of capitalistic economic development, they rather criticise the way the fruits of this development are distributed[31]. At any rate, it is not my concern here to discuss at length liberalism or socialism nor to find an answer to the problem of historical causation (this is rather a matter for philosophy of history). Yet, since this is not a paper in history nor in economics, but a philosophical work focused on the institutional critique of EA, I did not provide a full list of empirical evidence in favour of the Miller argument. Thus, I limited myself to reporting the empirical proofs against liberalistic mainstream economists he shows in his article Why Economists are Wrong About Sweatshops and the Anti-sweatshop Movement[32].


To better illustrate the practical implications of EA endorsing sweatshops, Iason Gabriel imagines a scenario where the rapid extension of sweatshops in a developing country is helping lift a large number of people out of poverty at the price of an increase in number of work-related accidents. If an effective altruist were asked to support an NGOs campaign that would improve the working conditions but reduce the number of opportunities for employment, however bad the working conditions might be, the effective altruist would choose not to finance the campaign[33]. Significantly, this is not only a hypothetical scenario. In fact, the ACIT (Academic Consortium on International Trade) worries that “if multinational corporations are persuaded to increase their wages (and those of their subcontractors) in response to what the on-going studies by the anti-sweatshop movement may conclude are appropriate wage levels, the net result would be shifts in employments that will worsen the collective welfare on the very workers who are supposed to be helped”[34]. However, Miller observes that, according to the calculations made by the economists Robert Pollin, James Heintz and Justine Burns there is evidence that this would actually not be the case[35].

Again, for EA what is Right comes before what is Good, in this case what are the basic (human) rights of the individuals can be sacrificed for the sake of the greater good. Yet, it might be emphasised that, although consequentialism is not intrinsically wrong and it is compatible with the appealing intuition that it is always permissible doing what leads to the best outcome, what makes applying strict consequentialism to this case mistaken is the impending uncertainty about the Good we are supposed to acquire: is that the actual most good? MacAskill would answer that, at least in this case, we can be almost certain that it is, partly because apparently there is nothing wrong nor exploitative with poor “choosing” the least bad option, partly because we have been assured by mainstream economists that the least bad option is a necessary evil, a stepping stone in view of a better future. The anti-sweatshop economist John Miller poses an important question:


[Given that] the “voluntary” exchange of labour for wages must be delimited by rules, collectively determined and obeyed by all, the relevant question is: What are those rules, and are any so basic that they should be applied universally? [(Miller’s question)] […]. For the most part economists, trained after all as economists and not political philosophers, have little to say on this matter other than to caution that outside of the condemnation of slavery, there is no universal agreement about the appropriateness of labour standards even when it comes to bonded labour and child labour[36].


The praise of sweatshops as “stepping stones” suggests that the answer MacAskill and many other effective altruists give to the Miller’s question seems to be that there cannot be universal standards when it comes to working conditions. This sounds very odd for a movement aiming to doing good.



Matteo Iammarrone

[1] William MacAskill, Doing Good Better, New York: Palgrave Macmillan, 2002, p. 15.

[2] MacAskill, op. cit., p. 15.

[3] Ibid.

[4]Helene Toner, Effective Altruism is a question, not an ideology, 16/10/2014, URL: https://forum.effectivealtruism.org/posts/FpjQMYQmS3rWewZ83/effective-altruism-is-a-question-not-an-ideology.

[5] Nonetheless, it is worth noting that not all effective altruists maintain consequentialist views. See for example: Ozymandias, You Don’t Have To Be A Utilitarian To Be An EA, 13/09/2016, URL: https://thingofthings.wordpress.com/2016/09/13/you-dont-have-to-be-a-utilitarian-to-be-an-ea/.

[6] MacAskill, op. cit., p. 159.

[7] MacAskill, op. cit., pp. 159-160.

[8] Ibid.

[9] The Paradox of Exploitation: A New Solution, Benjamin Ferguson, Department of Philosophy, Logic and Scientific Method of the London School of Economics and Political Science, September 2013, p. 11.

[10] Syme, T. (2019) “Charity vs Revolution: Effective Altruism and the Systemic Change Objection”, in Ethical Theory and Moral Practise, Vol. 22, Issue 1, p. 101.

[11] Ibid.

[12] MacAskill, op. cit., p. 160.

[13] Why Economists are Wrong About Sweatshops and the Anti-sweatshop Movement, John Miller, in Challenge, vol. 46, n. 1, 2003, p. 101.

[14] MacAskill, op. cit., p. 163.

[15] Ibid.

[16] Broi, A. (2019) Effective Altruism and Systemic Change, in Utilitas, Vol. 31, pp. 262-276.

[17] “I would say ‘socialdemocrat’ is a term that I am prepared to use”. See W. Pike. “Peter Singer, part 5 – politics and capitalism” YouTube, uploaded by W. Pike, 25/08/2011, https://www.youtube.com/watch?v=asvXb_LlLuU.

[18] Ibid.

[19] Ferguson, op. cit., p. 14.

[20] MacAskill, op. cit., pp. 159-160.

[21] Ibid.

[22] Ibid., p. 162.

[23] The argument of sweatshops as stepping stones on the path to economic development is presented on the website of the Adam Smith Institute. See Oliver Riley, “How Sweatshops Help the Poor”, Adam Smith Institute, URL=https://www.adamsmith.org/blog/how-sweatshops-help-the-poor, 20/03/2017. Accessed 18/09/2019.

[24] Simon Kuznets, “Economic Growth and Income Inequality.”, American Economic Review 45, no. 1, 1955, pp. 1‑28.

[25] Miller, op cit., p. 105.

[26] Miller, op. cit., p. 107.

[27] Miller, op. cit., pp. 107-108.

[28] Miller, op. cit., p. 106.

[29] The Cognitive Style of PowerPoint: Pitching Out Corrupts Within, Edward Tufte R., 2006.

[30] Determining what is a legitimate cause and what is accidental in a chain of events.

[31] Marx himself, as his dialectical view of history shows, recognised the social and technical progress of capitalism compared to his previous socio-economic stage (feudalism).

[32] Miller, op. cit.

[33] Gabriel, I. (2017) “Effective Altruism and its Critics” in Journal of Applied Philosophy, Vol. 34, No. p. 461.

[34] Miller, op. cit., p. 104.

[35] Ibid.

[36] See Why Economists are Wrong About Sweatshops and the Anti-sweatshop Movement, John Miller, in Challenge, vol. 46, n. 1, 2003, p. 103.

Articoli Correlati