In memory of Gerhard Øverland (1964 – 2014), teacher and friend.
Despite working an average six-day week and as much overtime as she can, sometimes coming home at midnight, Fatima struggles to earn enough money to support herself and her mother, who is unwell. Sometimes, Fatima chooses to go without food as she tries to stretch whatever little money she has until her next pay. Fatima, who has worked in [clothing] factories since her father died when she was 16, can only afford to live in a cramped two-bedroom apartment, which is shared with 10 people, including her landlord. They share a tiny kitchen and even smaller toilet and bathing area. Running water is available for one hour, just three times a day. One of Fatima’s roommates owns a thin single mattress, but Fatima sleeps on the concrete floor.
…
Even if big companies passed the entire cost of paying living wages to all workers on to consumers, … this would increase the price of a piece of clothing sold in Australia by just 1%.1
1. Introduction
If some part of a person’s or company’s profit depends on certain people being poor, does this have implications for the person’s or company’s obligations vis-à-vis these poor people, and if so, what are the implications?
Answering this question requires taking a stance on when someone counts as being poor. We employ an absolute notion of poverty, according to which a person is poor just in case she is unable, through reasonable work, to gain command over the resources and services she requires to lead a minimally decent human life (§3). It also requires understanding as precisely as possible what it means for profit to depend on poverty. We propose a counterfactual understanding of this dependence (§§2–3). As we shall see, one benefit of this approach is being able to properly consider poor people outside the workforce (§6).
To foreshadow, we answer: ‘yes, there are implications’. In particular, we defend the Poverty-Profit Principle (‘PPP’), which specifies conditions under which profit depends on poverty, is morally illicit, and must be redirected to those on whose poverty it depends (§§3–4).
Related conclusions might be reached in different ways, for example by arguing that when a person benefits from injustice (‘BFI’) she thereby acquires obligations to the victim(s) of that injustice. But the moral obligation we defend is distinct from that which arises from BFI because (1) the former always gives rise to a moral obligation when its conditions are met (but BFI only sometimes); (2) the former holds also where there is no injustice; and (3) the moral obligations are differently grounded. It also differs from exploitation in that (4) the latter, but not the former, is limited to those with whom some transaction takes place; and (5) these two obligations are also differently grounded. Appreciating the principle we defend therefore constitutes a genuine addition to our understanding of the moral landscape (§8).
We focus on a hypothetical company owned by a single person, which manufactures a single product, and whose employees are all paid the same wage. None of this is necessary, but it makes the presentation easier. The company moves production from a developed country to a region where many are poor. As a result, revenue increases. As we shall see, there are large sectors in which just such moves have very recently taken place on a massive scale (§2).
The owner of such a company may have responsibilities to poor people stemming from a variety of sources. A plausible principle says that if a person has the ability to prevent something very bad from happening at low cost to herself, she is morally obliged to do so (Singer Reference Singer1972, Reference Singer1999). This principle has immediate application to poverty. In developed nations, most adults can easily donate $50 a month to charity, which could secure an outcome equivalent to saving a child’s life about once every 20 months; or once a year at $85 per month.2 The avoidable death of a child is a very bad thing, and the cost, although not quite negligible, is low. So most adults in developed nations are plausibly obliged to donate at least this much. The owner will likely be affluent enough to incur obligations of this sort.
She may also contribute to poverty. Thomas Pogge argues that most affluent people in western democracies contribute to poverty by upholding institutions that foreseeably result in it (Pogge Reference Pogge2008).3 Perhaps the owner contributes in this way; her contribution may even be larger than that of most people. She may have obligations to the poor as a result. And she has still further obligations: to avoid damaging the environment, and to ensure that no children work in factories or supply-chains, for example.
Our question is whether she might be morally obliged to relinquish profit specifically because of the way in which that profit is related to poverty. To focus on this we assume that she has discharged all these other obligations, perhaps in part by donating to effective charities.4 To be clear, we do not assume that the owner has done all she is obliged to do to alleviate poverty. That depends on what’s at issue in this paper, namely on whether there are other sources of obligations to the poor.5 Our claim is that there are, and that one such is profit depending on poverty.
Finally, we assume that the owner makes the workers she employs better off than they otherwise would be. Making one’s workers better off is not enough to legitimise the continuation of a business practice, however, since one can wrongfully benefit others (Meyers Reference Meyers2004). Operations should continue, but the profit which depends on poverty must be relinquished, and redirected to those on whose poverty it depends.
We present two arguments for the PPP (§4). They will not convince everyone. But we think that the PPP captures an important moral obligation. In our experience, many are convinced by the Principle and by the arguments in its favour, and take it to articulate an important moral insight. We hope that the paper will be of value both to those who agree, and to those who do not. Perhaps some of the former can improve upon our reasoning, and make the arguments stronger. The latter will at least have a clear target to attack.
We have assumed that the owner has discharged all her ability- and contribution- based moral responsibilities. It would perhaps not be worth trying to establish the PPP if it weren’t possible for her to then have any money left to spend. It is plausible that each affluent person must sacrifice some of her income to (over time) prevent something very bad from happening. Just as the initial sacrifice is mandated, it may seem plausible that each additional sacrifice is mandated too, so that ability- and contribution-based obligations end up consuming all available funds.
Although ours is a world of widespread and desperate suffering stemming from poverty, we still think that discharging one’s ability- and contribution-based obligations will leave one with funds left over. Giving up a coffee per day is a small sacrifice. Giving away all one’s disposable income is not. It is a very substantial sacrifice, and would require strong justification, which is not obviously forthcoming. So the owner will have funds left over, and demonstrating an additional source of obligation relevant to her is worth our while.6
2. Poverty and Profit
Our imagined company has increased profits by moving to a region with many poor people. For this to be of relevance here, some part of the revenue increase must be attributable their poverty. Is this plausible?
To see that it is, note first the many indirect ways in which profit can be attributable to poverty. Laws and regulations govern a company’s operations; their stringency or otherwise impact on profitability. Stringent environmental and workplace safety laws and regulations, and strict enforcement, entail that a company typically spends more than it would otherwise: paying more to dispose of waste, expending more work-time erecting costly safety barriers, etc. In developing nations, laws and regulations associated with higher costs are often less stringent, and less stringently enforced, than in the developed world (World Development Report Reference Walton, Banerji, Cox, Diwan, Ghanem, Lindauer, Revenga and Rutkowski1995, p. 77). This fact is linked with poverty in several ways. Poor people may be unable to effectively agitate for stronger regulation, for example because of long working hours. Countries compete with each other for jobs for the poor, and are incentivised to provide lax regulatory regimes, and to set the minimum wage low (Emran and Kyriacou Reference Emran and Kyriacou2017). They are often ill equipped to enforce labour standards, in part because the informal sector is substantial (Meyers Reference Meyers2004, p. 329). Poverty is linked to profit in various indirect ways.
We take our conclusions to generalise to indirect cases such as these, and others.7 Since our focus is on the moral status of profit which is attributable to poverty it is helpful to focus on clear and direct cases. A particularly direct link between poverty and profit can be found in wages, working conditions, and benefits. Where many people are poor, there are many for whom accepting low wages, poor working conditions, and the absence of benefits, is the rational choice, because all other alternatives, including not working, are worse. Other things equal this makes a company which operates there (while selling to non-poor consumers) more profitable, because it can spend less (on wages, conditions, and benefits). Poverty is therefore part of the reason the company is more profitable than it otherwise would be. In such cases we say that part of the company’s profit is attributable to poverty.
To make this more precise we introduce some terminology. We assume, first, that x expresses that expenditure on wages which, for a certain company and period, maximises expected profits. If people would work for nothing, x would be zero. Since they (generally)8 will not, wages must at least be high enough to reliably attract enough workers to keep manufacturing going. In many cases, paying enough to retain workers over time is worthwhile, since efficiency increases outweigh the extra cost. (This may be counterbalanced by benefits of high personnel turnover, which hinders build-up of institutional memory of management promises, makes it more difficult for workers to organise, and so on). It may even be profitable to pay enough to instil feelings of gratitude or loyalty, if these are reliably enough connected with productivity increase. Other factors may also push x upwards. But when all factors are considered, and the only aim is profit, there is an answer to how much a company should spend on wages.
Determining what x is may be difficult. Companies surely often spend more or less than profit maximisation would dictate, because x is unknown. Such epistemic issues can be bracketed throughout, since our claims can be restated in terms of what x is believed to be. For ease of presentation we stick to the simpler language.
We assume, second, that y expresses the profit-maximising expenditure on favourable working conditions. Spending more reduces profits if nothing else shifts, but y can still be pushed up by a number of factors. In a warm country ventilation fans may increase productivity enough to outweigh the costs, and heating may have the same effect in a cold one. Finally, z expresses the expenditure on benefits which maximises profits. Should the company offer maternity leave and a pension plan? Should it subsidise or run transport, or health care? z is the profit-maximising aggregate expenditure on benefits.
Since x, y, and z express costs, the lower their value, the higher the profit, other things equal. We assume that companies typically set expenditure as close to x, y, and z as the regulatory framework allows. As noted, there’s a connection between poverty and the regulatory framework itself, including the minimum wage limit. This constitutes one link between profit and poverty. But the present point is that x, y, and z will typically themselves be lower in regions where poverty is widespread because, where many are poor, accepting low wages, poor conditions, and no benefits is rational for many, because all other alternatives are worse. Those who are already employed will find it hard to effectively negotiate for better conditions, since they are easily replaced, and have little time and energy after long and hard days of work (Meyers Reference Meyers2004; 329; Emran and Kyriacou Reference Emran and Kyriacou2017; 17). Holding other factors fixed, a company will therefore be more profitable where many are poor than where this is not so.
We are not claiming, of course, that all other factors are in fact always fixed in the real world. Countries without poverty can also compete. Yet, there are instances, and those are not uncommon, where profit depends on poverty.
By way of example, let us briefly consider the textiles, clothing, leather and footwear (TCLF) industries. In 2005, a ten-year phase out of import quotas was completed.9 This, together with the Global Financial Crisis, accelerated a massive change already in progress: away from manufacturing in high-income countries, and toward low-cost production in developing nations (ILO 2014, 1–2).
TCLF products are predominantly bought by the affluent.10 But production takes place elsewhere, and many developing countries are highly dependent on these industries. Clothing constitutes 88% of total exports from Haiti, 79% – 83% from Bangladesh, 52% – 67% in Cambodia, and 43% – 45% from Sri Lanka (ILO 2014; 8; Lopez-Acevedo and Robertson Reference Lopez-Acevedo and Robertson2016; Annex 2B). Clothing and textiles are estimated to account for nearly 15 million jobs in India, more than 11 million in China, around 10 million in Pakistan, nearly 4.5 million in Bangladesh, and around a million jobs each in Vietnam, Brazil, and Indonesia (ILO 2014; 9; Lopez-Acevedo and Robertson Reference Lopez-Acevedo and Robertson2016; Annex 2B).11
Reliance on better technology and more highly skilled workers makes the high-end segment of the sector less mobile, but the low-end segment moves quickly and easily (ILO 2014, 1–2). Governments therefore fear losing TCLF manufacturing jobs, and many set the minimum wage much too low for workers to avoid poverty. (It is US31¢/hour in Bangladesh, 50¢ in Indonesia, 51¢ in Vietnam, 66¢ in India, 74¢ in Cambodia, and 74¢ in China.) Moreover, workers are often paid less than the minimum wage, for example because falling short of impossibly demanding production quotas means wage penalties. Added to this is an absence of overtime pay, health insurance, and maternity leave; health and safety concerns (fire, exposure to chemicals); stressful and repetitive work; and violence, harassment, and discrimination (Emran and Kyriacou Reference Emran and Kyriacou2017, 10–12).
Poverty is widespread in TCLF production countries. The World Bank estimates that in Bangladesh around 28 million people lived on less than $1.90 a day in 2010.12 For Indonesia, that number was 38.5m. It was 3.65m in Vietnam; 261m in India; and 150m in China. Although highly influential, these numbers grossly underestimate poverty.13 The purchasing power afforded by $1.90 per person per day (in 2011 US) is clearly insufficient to escape poverty.14 On any reasonable standard of poverty, and certainly on the conception at issue here, the number of poor people in these countries is significantly higher.
All this makes it very likely that some portion of the (very substantial) profits from TCLF industries depends on poverty.15 Workers in low-end clothing manufacturing are easily replaced, and there is a large pool of potential workers for whom accepting the punishing conditions of TCLF industries is still the rational choice. TCLF is a large sector which most likely generates profit which depends on poverty.
3. The Poverty-Profit Principle
We have been making the descriptive claim that when a company operates where many poor people live, some portion of the profit is likely attributable to poverty. What is the moral import of this? We suggest the following normative principle, linking poverty and profit:
Poverty-Profit Principle (PPP)
For a company such that expenditure levels x, y, and z maximise its profit, and whenever there is a set of people — call it ‘the counterfactual set’ — such that
(1) all members of the set are poor, and
(2) if no member of the set were poor, then X, Y, and Z would instead maximise profit, and
(3) (X + Y + Z) > (x + y + z), then
(X + Y + Z) – (x + y + z) constitutes morally illicit profit.
According to the PPP, the difference between what the company would have had to expend on wages, working conditions, and benefits, if all the members of the counterfactual set were non-poor, and what it actually does expend, given that the members of that set actually are poor, is morally illicit profit.
(1) – (3) are sufficient conditions for profit being morally illicit. They are not necessary ones. Profit may be morally illicit for different reasons. Obligations vary accordingly. For example, profit is usually morally illicit if it depends on the employment of children. In such cases the profiting party cannot discharge its obligations by redirecting profit: ceasing employment of children is mandatory (at least if they will then go to school). Similarly, profit is morally illicit if it depends on practices which severely damage the environment. In such cases the company must modify its practices to stop the damage, or else stop them. When profit is illicit because the conditions in the PPP are met, however, the moral requirement is to redirect the illicit profit to another party. (We address the question: ‘to whom?’ in §6, below.)
The PPP links profit and poverty, and it is no small task to say what poverty is.16 Different notions may be appropriate in different settings. We have in mind an absolute notion of poverty, such that ‘[t]he poverty line has some absolute significance and to cross it is a change of some importance’ (Sen Reference Sen2009, p. 167).17 We are of course also interested in a notion which renders PPP plausible.18 Given this it is plausible that poverty largely consists in the inability through reasonable work to gain access to the resources (food, water, sanitation, clothing, shelter, etc.) and services (health care, education, protection, etc.) required to lead a minimally decent human life. More specifically, it is plausible that a person is poor if dignified and morally permissible work of reasonable duration and intensity, and in reasonable conditions — their own, or that of their provider — does not yield command over resources and services needed for a minimally decent human life.
This is a demanding conception of poverty. Faced with a situation in which a person did manage to command, through constant hard and long work, just enough to secure what’s required for a minimally decent human life, many would unhesitatingly judge her to be poor. However, our aim is to establish the truth of the PPP. Just as we focus on as clear and direct a link between poverty and profit as possible, we also focus on a clear and unambiguous notion of poverty itself, to make our case as strong as possible. And, just as we take our conclusions to generalise to less clear and direct links, we also think that they generalise to less demanding notions of poverty.19 Since not just any corporate move gives rise to morally illicit profit (§5), there is a threshold somewhere.
We emphasise the role that reasonable work plays in our conception of poverty. This ensures that cases where a person is strictly speaking capable of gaining the required income, but only by working in unreasonably dangerous, harmful, or unpleasant conditions (in unsecured mines, extreme temperatures, or damagingly noisy or dusty conditions, say), for very long hours, at extreme intensity, or carrying out immoral ‘work’ (for example, as an ‘enforcer’, who threatens violence), or work that undermines their dignity as a person (many, though not all, would regard prostitution as an example), still count as instances of poverty, as is, we believe, appropriate. Conversely, this conception also speaks to cases where dignified, permissible, and otherwise reasonable work is available, but the person chooses not to carry it out.20 On our conception of poverty, this person is not poor. The PPP is silent about profit which depends on her situation.
A lot more needs to be said. For one, what counts as work which undermines a person’s dignity depends on cultural and religious context. For another, our conception leaves open, as it must, what counts as unreasonably dangerous, harmful, or unpleasant conditions. And determining the nature of a minimally decent human life is itself an enormous and complex task which we make no attempt to tackle here.21 We recognise these challenges but will nevertheless assume that the notions in play are well enough understood for the discussion to proceed fruitfully.
4. In Favour of the PPP
Why believe the PPP? To see why it is crucial to keep the notion of poverty firmly in mind. A person lacking what’s requires for a minimally decent human life does not merely lack things she wants but can do without. She lacks things she cannot do without, and she suffers grievously for it. On the conception in play here, poverty entails deprivation and destitution.
With this in place, we offer two argument for the Principle. We do not presuppose a particular stance on the nature of morality (a consequentialist, Kantian, or contractualist view, for example). We take the arguments to stand on their own merits, and hope they will appeal widely.
4.1. Argument from Philosophical Conservativism
In our experience the Poverty-Profit Principle enjoys significant intuitive plausibility. Once the notion of poverty is elucidated as above, and once the principle has been explained and understood, many, though not all, have the intuition that it is true. Absent defeaters, intuitions provide justification for belief (Bengson Reference Bengson2015; Chudnoff Reference Chudnoff2011a, Reference Chudnoff2011b, Reference Chudnoff2013; Koksvik Reference Koksvik2011, Reference Koksvik2017). Absent defeaters, those who share the intuition are justified in believing that the PPP is true.
Some may be sceptical that people can have intuitions about the PPP, given its complexity. This is an unfounded worry: we routinely have true intuitions about matters at least as complex as the PPP.22 For some it takes a few minutes of considering the PPP before they have the intuition. But this is not uncommon in general, and does not negate that they really have the intuition with that content. As George Bealer has often noted, a typical reaction to first consideration of De Morgan’s laws23 is that nothing much happens at first, but then, suddenly, you ‘get it’, you just see that the transformations are valid. Similarly it may take a moment to grasp what the PPP says, but then its truth becomes apparent, at least to many.
Here as elsewhere appearances might be misleading. That does not make it rational to simply set them aside. Unless we find good reason to doubt them, those who have the intuition should take them at face value. As has been noted in a different context, this is simply ‘sensible philosophical conservativism’ (Pryor Reference Pryor2000, 538).
We claim there are no good reasons to doubt the Principle. Our defence of this claim has two parts. In §5 we clarify what the PPP does and does not say. The primary purpose is to avert misunderstandings. For example, if the Principle entailed that any profit-increasing relocation were morally illicit this would count against it. But the PPP has no such consequence. In §§6 – 7 we reply to further objections, for example to the claim that the PPP yields objectionable incentives. Together this constitutes a powerful case for the claim that are no good reasons to doubt the Principle. No such case can be conclusive: a better objection might be lurking in the wings. But until it makes its appearance belief in the PPP remains warranted for those who share the intuition.
That intuitions really do vary is a difficult but undeniable fact of life in philosophy.24 Some think this makes reliance on intuitions methodologically untenable (Weinberg Reference Weinberg2007; Swain, Alexander, and Weinberg Reference Swain, Alexander and Weinberg2008; Weinberg, Nichols, and Stich Reference Weinberg, Nichols and Stich2001). We do not agree. Our argument is made against the backdrop of several detailed recent developments of epistemological theory which holds that absent defeaters, intuition provides justification to believe what it represents (Bengson Reference Bengson2015; Chudnoff Reference Chudnoff2011a, Reference Chudnoff2011b, Reference Chudnoff2013; Koksvik Reference Koksvik2011, Reference Koksvik2013, Reference Koksvik2017). In none of its variations does that theory say that justification arises only absent inter-personal variance in intuitions — an unattainable standard for any source of justification. In what circumstances, then, does such variance constitute a defeater? This deep theoretical question is far from being settled.25 However, it is clear that the simple existence of inter-personal variance has no epistemic impact on its own. Given this background, those who share the intuition that the PPP is true, who are convinced by our arguments against potential defeaters (in §§5 – 7), and who do not believe we have overlooked anything significant, remain justified in believing that the Principle is true. This does leave those who have the intuition in a different epistemic position than those who do not. While uncomfortable, this is just what we should expect, so it is not on its own an objection to the argument here.
4.2. Argument from Moral Agency
Our second argument takes inspiration from one presented by Daniel Butt for the view that BFI gives rise to moral obligations:
The individual’s duty not to benefit from another’s suffering when that suffering is a result of injustice stems from one’s moral condemnation of the unjust act itself. … [T]aking our nature as moral agents seriously requires not only that we be willing not to commit acts of injustice ourselves, but … a genuine aversion to injustice and its lasting effects. We make a conceptual error if we condemn a given action as unjust, but are not willing to reverse or mitigate its effects on the grounds that it has benefited us. The refusal undermines the condemnation (Butt Reference Butt2007, 143, emphases ours).
Butt might appear to claim that the moral obligations arising from BFI are grounded in our moral condemnation of the original injustice; condemnation is what makes it wrong to refuse to relinquish the benefits. But this is clearly false: an agent who fails to condemn the original act isn’t thereby rendered blameless for not relinquishing benefits. A better interpretation is that the relevant ground is moral agency as such. We take Butt to say that moral agency in part consists in, and thus entails, aversion both to unjust actions and to their lasting effects. One cannot be a (‘fully formed’) moral agent without it.
In a world where no-one needs to be poor (§5), the fact that some people are may itself constitute an injustice. Butt’s argument then carries over directly. But even if not, a parallel argument can be run. There is inherent tension in on the one hand acknowledging that a state lacks what’s required for even a minimally decent human life, and on the other taking no umbrage to one person profiting from another being stuck in that state. The evaluation of a life as not even minimally decent cannot happily coexist with approval of the decision to profit from this very fact.26 The approval undermines the evaluation. Moral agency in part consists in, and thus entails, aversion both to people’s lives falling short of minimal decency, and to other people profiting from this very state of affairs. So, considerations largely parallel27 to those Butt gives above give us reason to accept the PPP.
The concept of a minimally decent human life is a moral concept. It is the moral evaluation of a person’s situation, and the moral acceptance of profiting from it, which we claim stand in tension.28
Should it trouble us that we use a moralised notion of poverty? No. Every notion of poverty relies on value-judgements. Poverty is essentially a concept not only of deprivation, but of bad deprivation (Wisor Reference Wisor2011; §1.8; see also Nussbaum Reference Nussbaum, Nussbaum and Glover1995). This strengthens the judgement of tension between the recognition of the moral badness of deprivation, and the approval of profiting from that very deprivation.
This line of thought will not convince everyone. Some will perceive no tension where we see it clearly. And there is only tension, and no incoherence, demonstrable with rigorous proof. Nevertheless, we think it is useful to point out, and make acute, the tension which we do see here, so that others may appreciate its force.
5. Clarifications
In this section we offer some clarifications of the PPP, with the aim of averting misunderstandings.
First, the PPP is restricted to worlds of moderate scarcity, in which it’s impossible to satisfy everyone’s demands, but possible to satisfy everyone’s basic needs (Rawls Reference Rawls1999, p. 110). We assume, as is common, that ours is such a world.29
Second, that the company moves is merely a dramatic device. Companies which for the first time set up operations, or which continue operations, are on a par with a company which moves.
Third, profit is illicit whether or not it is known to be so. What makes the profit illicit is that the dependence holds, not that it is known to hold.
Fourth, moving operations does not always gives rise to obligations under the PPP. Other things equal, a move from Norway to Australia will increase profitability, because the workforce is less affluent (let us assume). Such a company may incur obligations for other reasons, but the PPP does not deem the increased profit morally illicit.
Fifth, the PPP abstracts away from the details of the link between poverty and the conditions workers accept. This is one of its chief strengths. So long as the counterfactual dependence holds, further details do not matter for the profit being illicit. This allows some indirect connections to be captured, for example if poverty prevents unionisation and better wages. Moreover, there’s nothing special about connections via workers’ conditions. Principles which abstract further — capturing cases where the dependence goes via access to cheap parts, say — are also plausible.30
Sixth, we do not claim that further details never have moral implications. It may be worse to fail to give workers a decent salary than to fail to redirect illicit profit, for example. The PPP does not speak to this. Its conditions may be met in a variety of circumstances that otherwise differ significantly.
Finally, the most straightforward case of the PPP applying is when the workers in the company are themselves poor, and accept what they’re offered because all other options are worse. But things need not be this simple. Working for the company may bring all workers out of poverty, or they are not poor to begin with. The conditions may still be met. To see this, note that workers may accept the wages, conditions, and benefits offered because they know that others are ready to do so, and that they therefore will lose their job if they do not. In turn, those others may be poor. If those others would not accept the package were they not poor, they constitute the counterfactual set. Or these others may themselves not be poor, but have accepted worse conditions than those accepted by the workers in our company because they also know that they will lose their jobs if they do not, since still further people who are poor stand ready to accept them. Then this third group constitutes the counterfactual set. The company forming the ‘link’ here will itself be making illicit profit. However, here the point is that the illicitness of profit may ‘propagate upwards’ to companies which do not employ poor people, even to companies at significant apparent distance from poverty, so long as the required link is preserved and the conditions met.
6. To Whom?
When the conditions of the PPP are met we claim that the owner must redirect the illicit profit to someone else. But to whom?
If one focused only on the most straightforward case, in which the counterfactual set is exhausted by workers, one might conclude that illicit profit should always be spent improving workers’ conditions. However, the PPP can apply when workers start out poor but are lifted out of poverty by employment, and even where they are not poor to begin with. We take this to show that it is membership in the counterfactual set which entitles a person to a share of the illicit profit. Even where the counterfactual set is wholly constituted by workers it is not in virtue of being workers these people are correct recipients of the funds, but in virtue of being members of the counterfactual set.
This may seem untoward, since illicit profit may be thought to arise from underpaid labour.31 But keep in mind what grounds the obligation. It is not grounded in workers being paid less than some fair standard. It is not grounded in a transaction at all. It is profit depending on poverty — a morally bad state of deprivation — which grounds the obligation. Other obligations may also obtain. But the moral wrong of (say) paying workers less than their fair share is a different matter than the wrong of profiting from poverty. This point is illustrated by the sets of people to whom obligations are owed not necessarily coinciding, but originates in the deep differences in how the obligations are grounded.
When profit is illicit because the PPP applies, it should, other things equal, be redirected to all and only the members of that set. If workers remain poor after being paid they are in the set, and may even exhaust it. But if the workers are not poor after being paid it is to those that are poor the money should go. Typically, both workers and many others are poor. There is then strong reason to ensure that the money does not go only to those ‘lucky’ enough to have secured a job, but also to the less fortunate poor.32
A consequence of our view is that children, the elderly, and the infirm may not receive redirected profit, even if they are among the worst off, since profit may fail to depend on their poverty. Perhaps some workers only accept because they have dependents to care for. But children, the elderly, and the infirm will only be members of the counterfactual set if people are willing to work to support them. Relatedly, when profit is directed to the counterfactual set, it is not guaranteed go where it can do the most good.
We think these consequences should be embraced. The PPP provides a pro tanto moral reason to act. There are other moral reasons. As Kagan notes, pro tanto reasons always have weight (in contrast to prima facie reasons, which may merely appear to be reasons) (Kagan Reference Kagan1989, p. 17). But they need not be decisive, and can be outweighed. The obligation captured by the PPP gives reason to redirect profit to members of the counterfactual set, but if much greater good can be secured in a different way there may be all-things-considered moral reason to do that.
This is not peculiar to the PPP. For instance, the principle of contribution says that other things equal one has stronger reason to direct funds to those to whose suffering one has contributed than to others. However, if you cause damage to someone’s car and therefore owe him compensation it may still be incumbent on you to use all available means to help someone in much greater need.
7. Further Objections, and Replies
7.1. Perverse Incentives
Another type of objection alleges that the PPP is false because it yields perverse incentives.
Suppose a company intends to move production, and is considering two destinations: ‘BNP’, where none are poor but many are barely non-poor; and ‘P’, where many are poor. Bracketing the PPP it has an incentive to move to P. Profits will be larger, since the poor will accept worse conditions than the barely non-poor. The PPP dictates that this difference must be relinquished, so the company no longer has an incentive to move to P over BNP. But it ought to, since providing work to poor people constitutes a greater good than to the barely non-poor. Moreover, if it is observed to make greater profits in P than elsewhere, other companies will also be incentivised to move there. The resulting competition will benefit the poor, a benefit lost if the PPP is heeded. So the PPP yields perverse incentives, or removes good ones, and must be rejected.
Contrary to the objection, however, the company does have a reason to move to P rather than to BNP. Providing work to poor people, and helping (other) poor people by redistributing profit, creates a more significant benefit than the alternative, exactly as the objection notes. That constitutes a reason to move to P over BNP. It is not a self-interested incentive, and the company may not be moved by it. But the mere possibility of a situation in which an agent, cognisant of one moral principle but ignoring others, fails to be correctly incentivised, does not constitute an objection to the truth of that principle, though it may speak against its codification (on which more shortly). Such situations can be generated for all (non-total) moral principles, and so would not leave any left standing.
A related objection claims that the owner must be allowed to derive profit from poverty, since employing poor people is associated with increased risk, or social stigma, so that no-one will do it without the extra benefit.33 There are two interpretations of this objection. On the first it is a strengthening of that just discussed: not only will the company lack motivation to move to P, it will have motivation to move to BNP instead.
However, there is little reason to think that hiring poor people over barely non-poor people adds notable risk, or that there is significant additional social stigma. Beyond this our reply is the same: we are not concerned with how to legislate or regulate but with the truth about our moral obligations, and the company still has a strong moral reason to move to P over BNP.
On the second interpretation, the risk and social stigma has moral weight, and genuinely counts against the reasons the owner has for moving to P over BNP. Insofar as there really is increased risk we agree that this must be taken into account, but hold that it must be solved in some other way — through a mandatory collective insurance scheme, government backed security, or what have you — since it does not make it acceptable to profit from morally bad deprivation. The owner might have a claim against other non-poor people, but this does not bear on her obligations to the poor.
Stigma from employing poor people will hardly stem from poor people themselves, so this component runs together two moral relations: what the owner owes to the poor, on the one hand, and what other non-poor people owe her, on the other (Karnein Reference Karnein2014). The owner not being afforded adequate status or respect by other non-poor people has no bearing on her obligation to the poor.34
In general, the possibility of incentive structures such as we have been discussing at best shows that we can have reason to allow legal or regulatory requirements to vary from moral ones. Perhaps the PPP should not be reflected in regulation at all; this holds for many moral principles. Even if it should be it is an open question as to how: perhaps regulation should only require relinquishing some portion of the illicit profit, so that beneficial incentive structures are retained for actors with selective moral attention.
That we might have reason to let regulation imperfectly reflect moral truth is a familiar point. It does not bear on the truth or falsity of moral principles. And we are concerned with moral truth, and not with which ‘conduct-guiding structures’, of laws, values, norms, or what have you, we should try to effectuate (Pogge Reference Pogge1992). That is an interesting question too, but not our concern here.
7.2. Over-demanding
We are committed to a morally significant threshold, such that where a person falls with respect to it impacts on our obligations to them. An objection takes point of departure in this feature of the account.
Consider a poor slum dweller with a fabulous business idea which exploits an empty niche in the marketplace. She skilfully implements it, hires poor people, benefits her workers, and makes good money. The PPP seems to say that she may not enrich herself further than to barely escape poverty unless she also lifts all her employees out of poverty; indeed until she ensures that the counterfactual set is empty. That may seem unduly demanding. It was her idea, after all, and one might think she should be able to reap a greater reward than this for having the idea, and for skilfully implementing it.
Our answer proceeds in two steps. Note first that only a small portion of a company’s profit will usually depend on poverty.35 A company may be profitable because it exploits an empty niche, sources raw materials cheaply, employs efficient methods of production, and for many other reasons. Moreover, a worker typically adds greater value to a product through her efforts than the difference between the conditions she does accept and those she would have accepted had the counterfactual set been empty. So a company will nearly always be able to relinquish illicit profit while having significant profit left over. The case as described rarely obtains.
But if this is, as per very unusual, not the case, then the poor entrepreneur does have an obligation to relinquish the profit. She is no more permitted to reap financial reward from other people’s lack of what’s required for a minimally decent human life than anyone else is. That is a demanding conclusion. Poverty is an extremely serious condition. Demanding conclusions are to be expected.
It bears noting that the entrepreneur’s profit may completely fail to depend on poverty. In that case, the PPP does not speak against her keeping all of the profit for herself, though other considerations might. This does not undermine the plausibility of the PPP, though it does say something about its scope.
7.3. Unfair Share
A related but distinct objection proceeds as follows. We have restricted the PPP to worlds of moderate scarcity. But, says the objector, if in such a world there is poverty even after the owner has discharged her assistance-based and contribution-based obligations, that must be because other agents have failed to discharge theirs. The owner is being asked to do more because others have done less (than they ought to have). This is unfair, particularly since the owner has discharged her other obligations to the poor, and benefits poor people by offering them jobs.36 She is doing a lot. We cannot demand more.37
For this objection to have force, the owner’s obligation under the PPP would have to stem from a collective duty to eradicate poverty which first fell to a group to which she belongs, and which had then been divided into ‘fair shares’, including the owner’s. But that is not so. The source of the obligation is the relationship between poverty and profit. There would be no obligation if no-one were poor, but that’s because the owner would then be unable to generate illicit profit to begin with.
The objection relies on a further mistaken assumption, namely that the notion of a ‘fair share’ even applies to ability-based and contribution-based obligations. Holly Lawford-Smith (Reference Lawford-Smith2015) and Stephanie Collins (Reference Collins2017) have shown that this notion does not apply to ‘the affluent’, the only candidate group of relevance here. Groups like ‘the affluent’ lack the required structure to act. Since they cannot act, they cannot be obliged to act. So there is no ‘fair share’ that falls on the owner in virtue of her membership in the class of affluent people.
Moreover, even if the notion did apply, the objection would still fail, for reasons Collins gives. Anja Karnein (Reference Karnein2014) has argued that a ‘non-complier’s duty to fellow actors’ is distinct from ‘a fellow actor’s duty to third parties’. The unfairness objection only works if ‘upholding fairness between actors … is more important than remedying harm done to others’ (Collins Reference Collins2017, 582). This is not always the case: unfairness can be negligible relative to harm.38 Whether it is so in a particular case depends on ‘the severity of the intra-group unfairness, as against the severity of the harm to third parties, as against the cost of remedying that harm’ (582). Suppose I take on ‘your’ burden to donate $50 a month, and save a child’s life every 20 months. Ex hypothesi I suffer an unfairness, but this is nowhere near significant enough to absolve me of my obligation, given the importance of an innocent child’s life. Similarly, even if the owner did suffer an unfairness from other non-poor people, it would be nowhere near significant enough to absolve her of hers.
8. The PPP, Benefiting from Injustice, and Exploitation
The PPP purports to capture a source of moral obligations to poor people, so it’s interesting to consider how it relates to other such sources, in particular benefiting from injustice (‘BFI’) and exploitation. We now argue that the source of obligation captured by the PPP is distinct from both of these.
We have no interest in the merely verbal dispute (Chalmers Reference Chalmers2011) of whether what the PPP captures should be called ‘BFI’, or ‘exploitation’. We also take no stance on the (substantive) question of whether, when moral theory is completed, PPP will fall within one of the categories picked out by those names. This depends, among other things, on the nature of moral kinds, a question well outside the scope of this article. But we do claim that as those categories are currently understood, the PPP, as a source of obligation to the poor, differs from both. So we do not rely on, and need take no stance on, the moral status of either BFI or exploitation.
Where poverty is preventable, it’s not unnatural to think that to be poor is to suffer an injustice. One might thus think that the obligation we’re after is better captured by a principle which says that when one benefits from an injustice one incurs obligations to its victim(s). After all, to profit is to benefit.
Daniel Butt is an early proponent of a view of this kind (Butt Reference Butt2007, Reference Butt2009, Reference Butt2014). He defends the following principle:
Agents can come to possess obligations to lessen or rectify the effects of wrongdoing perpetrated by other agents through benefiting, involuntarily, from the wrongdoing in question (Butt Reference Butt2014, p. 338).
Butt adopts Miller’s ‘connection theory’ framework.39 The central claim is that several different connections to a bad situation40 can be relevant to assigning responsibility to assist: causal and moral responsibility, capacity, and community. Butt adds BFI as a further connection, a claim Miller later accepts (Miller Reference Miller2007, pp. 102–3). But both agree that none of these connections is guaranteed to give rise to responsibility. We have to consider each case on its merits, usually several connections are relevant, and judgement comes down to intuitions about which connection weighs more heavily. Having benefited from injustice may easily fail to give rise to any obligation whatever.
That benefiting from injustice is sometimes but not always a source of moral obligation is widely accepted in the literature concerning BFI (Goodin and Barry Reference Goodin and Barry2014; 364–65; Anwander Reference Anwander2005; 40–41; Haydar and Øverland Reference Haydar and Øverland2014, 351–52; Caney Reference Caney2010; 210; Lawford-Smith Reference Lawford-Smith2014; 396–400; Goodin Reference Goodin2013; 483, 488; Barry and Wiens Reference Barry and Wiens2014). This stands in principled contrast to our claim. We hold that when conditions (1) – (3) are fulfilled the PPP always generates a moral obligation. It’s a pro tanto obligation which may be outweighed, but it is always there, and it always has weight.
Second, while it’s not unnatural to think that to be poor is to suffer an injustice, we do not rely on this assumption. The PPP is plausible as it stands, with no mention of injustice. Its plausibility is not diminished by the stipulation that a particular case of poverty does not constitute an injustice. The intuition remains strong, and it remains plausible that moral agency in part consists in aversion to profit which depends on morally unacceptable deprivation.
Finally, what grounds the moral obligations from BFI is a relation between an act of wrongdoing, on the one hand, and either the event of receiving a certain benefit, or the state of retaining and enjoying it, on the other. By contrast, what grounds the obligation captured by the PPP is a relation that in the first position has a state of morally unacceptable deprivation. How that state came about does not matter.
To sum up, (1) the PPP always gives rise to a moral obligation which has weight, contrary to a universally accepted feature of BFI. Moreover, while BFI takes point of departure in an injustice, (2) the PPP also holds where there is no injustice, though still morally unacceptable deprivation. And finally, (3) the obligation captured by the PPP is differently grounded than the obligation arising out of BFI. The PPP is thus not reasonably regarded as a species of BFI; it captures a distinct source of moral obligation. Consequently, we need not defend BFI as a source of moral obligation. If it is, some arguments carry over, but they are not needed to establish the PPP. The case for the Principle is independently plausible.
Exploitation is often discussed in connection with sweatshops; factories where workers toil long hours in poor conditions. An important question is what working-conditions and wages a multi-national company should, morally speaking, offer sweatshop workers. Some claim that, when determining the level of wages, conditions, and benefits a company should offer, ‘[t]he appropriate test is not whether the wage reaches some predetermined standard but whether it’s freely accepted by (reasonably) informed workers’ (Maitland Reference Maitland1997, p. 607). Others argue that there are substantial limits to how low wages can be (Meyers Reference Meyers2004). Since there is clear overlap with the cases and questions which have been our focus, this source of obligation may be thought to already cover the ground we have been traversing.
Øverland argues that, for a case to qualify as exploitation, the exploited party must have available a given option (a status quo they can allow to continue); an exploitative option must be presented; their acceptance must be voluntary; and the transaction must be beneficial to both parties (Øverland Reference Øverland2013, 323–24). This shows that the PPP is distinct from exploitation, since these conditions needn’t be met for the members of the counterfactual set. Many and possibly all the members of the set are not made any offer whatever.
Øverland’s is of course not the only account of exploitation worth considering, but the point generalises widely. For example, Allen Buchanan argues that ‘to exploit a person involves the harmful, merely instrumental utilization of him or his capacities, for one’s own advantage or for the sake of one’s own ends’ (Buchanan Reference Buchanan1985, 87). The non-working members of the counterfactual set do not qualify as being exploited under this definition, since a) they are not plausibly utilised by the company’s owner, or, if they are utilised in some attenuated sense of that word, b) that utilisation is not harmful. Nancy Holstrom’s position is that ‘[i]t is the fact that the [capitalist’s] income is derived through forced, unpaid, surplus labor, the product of which the producers do not control, which makes it exploitative’ (Holstrom Reference Holstrom1977, 359). Again, this does not apply to non-working members of the counterfactual set: they are not labouring for the capitalist. In general, a common thread is that some transaction must take place between the parties for exploitation to occur. By contrast, part of our point has been that a weaker relation, namely counterfactual dependence of profit on poverty, suffices to generate an obligation, even in the absence of any prior transaction between the agent with the obligation and those to whom it is owed.
With the possible exception of the limiting case, when the counterfactual set is exhausted by the workers, the members of the set are not (all) exploited by the company under consideration. It is not in virtue of entering into a financial relationship with the company, or in virtue of participating in an exchange with it (Meyers Reference Meyers2004, p. 328), that the people in the set are morally relevant, but in virtue of the profit’s dependence on their poverty. The obligation captured by the PPP is thus distinct from obligations not to exploit. Moreover, our argument shows that exploitation is not the only, and may not be the most important, moral problem that arises in sweatshops and similar circumstances.
Finally, the obligation captured by the PPP, and that not to exploit, are differently grounded. The latter is grounded in a relation between the act of offering or continuing to offer exploitative conditions, on the one hand, and the event of making an (overly large) profit, on the other. By contrast, the PPP is grounded in a relation that in the first position has a state of morally unacceptable deprivation. How it came about does not matter.
To sum up, the moral obligation captured by the PPP differs from that which arises from exploitation in that (4) the latter, but not the former, is limited to those who work for the company, or those with whom some transaction takes place, and (5) in that the two obligations are differently grounded. The moral obligation captured by the PPP is thus not reasonably regarded as of a kind with those that arise out of exploitation. It captures a distinct source of obligation. In arguing for the PPP we therefore need not defend a particular stance about exploitation. The case for the PPP stands independently.
9. Significance
We end by considering a final objection, namely that the Poverty-Profit Principle overgeneralises. In a world as interconnected and with as high incidence of poverty as ours, so the thought goes, it might be that everyone’s profit (or income, or savings, e.g. due to paying less for consumer products) depends, to some extent, in some way or other, on poverty. That would yield the result that some part of everyone’s profit (income, savings) is morally illicit; an eye-popping prospect, to say the least.41
We say that what matters is whether the dependence relation between profit (income, savings), on the one hand, and poverty, on the other, in fact obtains, and if it does, the consequence should be adopted. This possibility merits much greater scrutiny than we can give it here. It is one of the reasons why the PPP is important.
Insofar as a person’s or company’s profit (income, savings) depends on people’s being poor, however far removed from poverty the person or company might initially seem to be, that profit (income, savings) is morally illicit, and should be redistributed to the people in the counterfactual set. There will usually be significant epistemic problems associated with working out whether a given person’s or company’s profit (income, savings) actually is counterfactually dependent on poverty, and, if so, how large a proportion of it is. But bracketing such challenges, if the dependence holds, then the conclusion, although uncomfortable, is correct.
Acknowledgements
This paper has been in the works for a long time, and many people have helped along the way. It was presented at the ANU Postgraduate Workshop at Kioloa in 2011; at the University of Tasmania, and the ‘Responding to Global Poverty’ workshop in Gerhard’s honour at the University of Oslo, both in 2015; and at the Seminar on Practical Philosophy at the University of Bergen in 2016. I am grateful to those present for comments and questions. For detailed comments, in writing or in person, I am indebted to Leon Leontyev, Robert Kirby, Holly Lawford-Smith, Stephanie Collins, Garrett Cullity, and Nicholas Southwood; and to two anonymous referees for this journal. For discussion in other contexts thanks go to Christian Barry, Bashshar Haydar, Patrick Tomlin, Ole Hjortland, Jesse Tomalty, Espen Gamlund, John Cusbert, Rosa Terlazzo, Dirk Baltzly, and Elaine Miler. Most of all I am grateful to Gerhard, who cared not one iota about where ideas came from, but very deeply about the ideas themselves. This article has changed a lot since he last saw it. I hope he would have been proud.
Disclosure statement
No potential conflict of interest was reported by the authors.
