I feel like there is a lot of exploitation in the world. When I buy clothes, I worry that they have been made by exploited workers, labouring in appalling conditions in sweatshops in developing countries. When I use my mobile phone, I worry that the coltan that is used to manufacture the chips has been sourced from exploited workers in conflict zones, and that the phones themselves have been assembled by exploited workers in large factory complexes somewhere in Asia. Of course, I still buy the clothes and use the phone (like pretty much everybody else). So the question arises: should I worry about the exploitation?
To answer that, one needs to have a clearer sense of what exploitation entails and when it is morally objectionable. Providing such a sense is the purpose of this blogpost. Following a recent article by Steve Wilkinson, I am going to attempt three things. First, I’ll give a general definition of exploitation, highlighting two conditions that need to be satisfied in order for something to count as an instance of exploitation. Second, I’ll motivate the two conditions by considering some thought experiments that suggest that both are necessary elements of the definition. And third, I’ll discuss a range of problem cases that help us to better understand the wrong involved in exploitation.
Given that Wilkinson’s article deals with commercial surrogacy, and given that commercial surrogacyis one of myown interests, much of the ensuing analysis uses surrogacy-related examples to explore the contours of exploitation. But the goal is not to provide an analysis that is limited solely to those examples. The goal is to provide something of more general significance.
1. Defining Exploitation
Your friend recently started a job as a chef. She complains that her boss makes her work long hours; that she does the lion’s share of the cooking every evening; and yet the head chef makes much more than she does. She says she feels exploited. Is she right?
Possibly. The complaint and the scenario seem like standard everyday examples of exploitation. I know I have had friends make similar complaints to me. I agree that their working conditions sound unpleasant, but what is it that makes something unpleasant or undesired an instance of exploitation? If your friend accidentally burnt herself while cooking one evening she wouldn’t say that she felt exploited, would she? Something bad happened to her, for sure, but it didn’t have the properties we would typically associate with exploitation. And what if her boss held her finger against a hot stove and deliberately burnt it? Would that be exploitation or would it just simply be a serious form of assault?
This isn’t a purely semantic question. The labels are important because they help us to distinguish between distinctive kinds of wrong. Exploitation may warrant a different response than assault. Drawing the line between them might be important when it comes to practical, policy-related concerns.
Wilkinson proposes a two-part definition of exploitation (the following is taking with modifications from Wilkinson 2015, 3):
Exploitation: Person A can be said to be exploited by Person B (and/or C, D, E….) if:
(a) person A derives (or is at the risk of deriving) an unfairly low level of benefit and/or suffers an unfairly high level of cost or harm from their interaction/transaction with B;
(b) person A’s consent to the arrangement is defective or invalid in some way.
As Wilkinson points out, condition (a) in this definition is the most intuitive. It seems like this is the factor at play in your friend’s description of her worklife as being exploitative. Her point is that she is not being properly rewarded for the amount of work she is putting in. Someone else (the head chef) is free-riding on her hard labour. She is receiving an unfairly low level of benefit.
Condition (b) is the tricky one and one that is often absent in everyday examples. It is not clear, for instance, that your friend’s consent to work under those conditions was invalid. We don’t know all the details, but suppose she willingly took on the job and knew (to some appropriate degree of specification) what it would entail? According to Wilkinson’s definition, if that were the case we would not be allowed to say that she has been exploited. She might be unfairly paid and deserve more money, but she is not exploited.
2. Do we need both conditions?
Is this right? Do we need both conditions in order to have exploitation? Wilkinson thinks we do. We can illustrate his thinking using some thought experiments. First, imagine a scenario in which condition (b) is met but condition (a) is not:
Non-consensual lottery winner: Paula woke up one morning to discover that she had been randomly selected by the government as the winner of a lottery worth over €1 billion. She is told that there are no strings or conditions attached. She will receive the money immediately. She can spend it however she sees fit. And her identity will be protected. The only thing she cannot do is reject the money.
In this case, Paula has received a massive benefit without her consent. There are no (obvious) downsides or negatives (apart from perhaps the psychological adjustment involved). Would we be inclined to say that Paula is being exploited here just because she received the money without her consent? (I should note, in passing, that the scenario is not all that uncommon: wealthy heirs are effectively in this position).
I would argue that Paula is not being exploited in this scenario. Maybe something less than ideal is happening to her — maybe it would be better, all things considered, if she had been given the opportunity to consent to the arrangement — but it doesn’t feel right to equate her predicament to that of the poorly paid factory worker or a trafficked sex worker. This suggests that condition (a) is essential to exploitation.
Now imagine a scenario in which condition (a) is met but condition (b) is not.
Furniture Fire Sale: John recently got a new job in a new country. He cannot bring the furniture and equipment he has in his current flat with him when he goes. He needs to get rid of it. He estimates that the actual current market worth of his furniture and equipment is somewhere in the region of €1500. But given his predicament he sells it to a friend for €150.
The price is unfair. John could have got a lot more if he had sold through the typical channels. But John is willing to accept much less due to his current situation. He needs to rid himself of unnecessary stuff before he leaves the country. He is consenting to the below-market-rate price, fully aware of the fact that he could have got more. Given this, it is difficult to say that he is being exploited. An unfairly low price, accepted with full consent, simply doesn’t feel like a candidate instance of exploitation
This is a general point, one that Wilkinson is keen to emphasise. When condition (a) is satisfied but (b) is not, you may really be dealing with a situation involving generosity, bad luck or negligence. The general point is that the reasons why somebody accepted a low offer are always going to be relevant when determining whether or not it counts as exploitation. Hence, why condition (b) is essential. It tells us that there is something importantly defective or invalid about those reasons.
3. Problem Cases: Package Deals and Background Conditions
Now that we have basic sense of what exploitation consists in, let’s consider some problem cases. We’ll use transnational gestational surrogacy (TGS) as our case study. It has become common for affluent couples in developed nations to use women in developing nations (most noticeably India) as paid surrogates. Questions have arisen as to whether this practice is exploitative. Many feel that it is. On a previous occasion, I looked at an extended argument in favour of this view.
Today, I want to use a slightly artificial example involving TGS to probe some of our intuitions about exploitation:
Paid Surrogacy: Riya lives in the Anand province of Western India. She has two children and a husband. She normally works in a factory, but she was recently offered $3,000 (USD) to become a surrogate mother for an Australian couple. This is more than ten times the annual salary she can earn in the factory. She meets with other women who have gone through the process and she is reassured by their stories: they tell her that although there are some risks and costs associated with the process, the benefits outweigh those costs. She agrees to the arrangement.
The example is artificial because it presumes that Riya enters into the arrangement with a decent understanding of what it entails. It’s not obvious that this is the case in practice. This is something I discussed in a previous post. It is not artificial in terms of the sum of money (and its relative value). This is broadly in line with reality. This makes the example interesting. If we followed the two conditions in Wilkinson’s definition, it would seem like there is no obvious exploitation taking place. Riya is giving informed consent to the agreement (or as informed as you can ever be when you choose to become pregnant). And she is not receiving an unfairly low level of compensation. On the contrary, she is receiving a very high level of compensation.
Is there, nevertheless, something exploitative about the transaction?
Some people think there is. They target the consent requirement in their arguments. They claim that one thing you need for consent is the absence of manipulation/coercion. They argue that the choice with the Western couple confront Riya is structured in such a way as to render her consent invalid due to manipulation or coercion.
There are two ways of cashing out this thought. The first is to argue that the offer is too generous. The level of compensation is such that she is no longer able to refuse. The Western couple are taking advantage of her relative poverty and their relative affluence to offer more money than she could hope to earn in any other context. She is thus being manipulated into giving her consent by an excessively high compensation.
This is an interesting claim, but if successful it would force some change in condition (a). After all, it is saying that exploitation can exist even when there is nothing unfairly low about the compensation. On the contrary, it is saying that it can exist when the compensation is too high and this is overriding her ability to meaningfully consent. The problem really has to do with the power disparity and the way this is leveraged. So maybe condition (a) should speak to relative asymmetries of power instead of unfair levels of compensation.
That amendment would still leave open two question: can you meaningfully consent to a very high compensation? And does it matter what the compensation is for? If a billionaire offered you a million dollars, no strings attached, it seems like you could meaningfully consent to it. So maybe the problem is when the high level of compensation comes as part of a package deal that would otherwise be unacceptable? What the woman really wants is the money, but in order to get the money she is required to provide the gestational service. She wouldn’t consent to the gestational service without the money. The consent to the package deal is, therefore, tainted by the fact that there would be no consent to provide the service itself. The package deal is thus the problem.
Or so the logic seems to go. Janet Radliffe Richards has critiqued arguments of this sort. She points out that they seem to commit us to the view that if you don’t desire X in itself you cannot consent to it all things considered. But this seems absurd. It would call into question a number of fairly standard decisions. For instance, you may not like being jabbed with a needle, but you can appreciate the fact that being jabbed with a needle sometimes has beneficial medical effects. Does the fact that you would not consent to being jabbed with a needle in and of itself mean you cannot consent to it all things considered (given its benefits)? Surely not. Why is it different for the surrogate?
This brings us to the other way to cash out the thought that Riya is being exploited in the example given above. Perhaps the problem is that her perception of her choices is being unfairly narrowed by her (relatively) less fortunate background conditions? The $3000 is attractive to her because she is relatively poor and has few other employment options. She can work in the factory for less money or become a surrogate. Neither prospect is ideal. Becoming a surrogate is really only the better of two evils. She is thus exploited because her background conditions make her options limited. This is presumably known to the affluent Australian couple and is what drives the entire trade.
Now, there is an important qualification to add here. Bad background conditions by themselves will not render consent invalid and give rise to exploitation. It matters greatly what the cause of the background conditions is. If your limited options are caused by brute luck, then you are probably not being unfairly taken advantage of if someone gives you an offer you can’t refuse.
Suppose tomorrow you are diagnosed with a serious form of cancer that has no known cause. The oncologist tells you that you have two options: (a) do nothing and die within 6 months or (b) take an unpleasant and expensive course of chemotherapy and have a reasonable chance of living up to five years. If you choose (b), is your consent invalid? Is your doctor exploiting you? Again, surely not. The brute luck makes the narrowing of choices morally unobjectionable. Contrast that with the classic case of the highwayman who waylays your stagecoach, holds you at gunpoint, and gives you a choice (a) your money or (b) your life (and presumably your money as well)? Consent is invalid in that case due to clear external coercion. It’s the morally objectionable coercion that limits the range of options.
The question then becomes where on spectrum between brute luck and clear external coercion does the case of the Indian surrogate lie. You could argue that her poverty is just brute luck, but that doesn’t seem right. It is at least partly the result of deliberate human decision-making. Wilkinson quotes another author (Baylis) in his paper. Baylis argues that Riya’s situation is partly caused by the fact that the Indian government has pursued economic growth through medical tourism, at the expense of social justice, and without developing a robust regulatory regime for medical tourism. This might tip her predicament closer to the external coercion end of the spectrum.
Wilkinson agrees that this might make the scenario exploitative. But he doesn’t think this means that paid surrogacy should be banned. Exploitation only gives us a reason (perhaps a strong one) for objecting to a practice. But that reason must be weighed against others that might stand in favour of the practice, along with the potential costs and benefits of banning or restricting the practice. He suggests that when you conduct this weighing exercise, the rationale for a ban becomes much less persuasive.
Okay, I’ll leave it there. To briefly recap, a good first pass at a definition of exploitation focuses on two conditions (a) unfairly low benefits and (b) absent or invalid consent. This definitions tracks well with our intuitions, but there are problem cases. One such case is that of the international surrogate who actually receives a very high benefit for providing a service. The suggestion is that there is may still exploitative about those transactions because of the relative asymmetry of powers between her and the affluent couple who procure her services. This requires her to provide consent to a problematic package deal under morally objectionable background conditions.