This is a crosspost from my blog
Reparations
Many critics of global inequality argue that our current system isn’t just broken, it’s historically unjust. Decolonial thinkers say that combating this inequality isn’t just about helping the poor; it’s about repairing centuries of exploitation. They argue that reparations, not just aid, should be at the heart of our response.
Why? Because Western nations built vast amounts of wealth through colonial slavery and the exploitation of Sub-Saharan Africa. And the legacy of that exploitation hasn’t disappeared. It lives on today through unfair trade deals, crushing debt burdens, and subsidies that distort global markets to the disadvantage of African farmers.
In this view, justice is about paying back what was taken, and thus reparations are a form of compensatory justice: if you benefit from someone else’s suffering, you owe them something. It’s not about victimhood; it’s about taking responsibility.
As some scholars in development studies put it:
[In order] to redistribute global wealth and end growing inequality reparations have to be made. The North has to take responsibility for the wealth it has built over centuries at the expense of the South. These reparations are not about compensating victimhood, rather they constitute the radical claim of a growing number of individuals and movements globally for social justice.
Targeting tax havens and striving towards a just taxation worldwide would be a first, although still very modest, step towards a financial model for reparations. Thus, if we abolish [the position of European Union Commissioner for Development], why not replace it with a commissioner for reparations?
Reparations are, in theory, not just about handing over money. If other kinds of policy can meaningfully undo the harms of colonialism, they count too. Still, in practice, financial compensation is by far the most popular proposal.
Writers like Ta-Nehisi Coates and philosophers like Olúfẹ́mi Táíwò have argued for concrete reparation payments. Closer to home, activists in the Netherlands calculated how much a single enslaved woman would be owed by a modern-day foundation, and Belgian politicians discuss compensating for the legacy of colonization in Congo. Although small steps have been taken, full reparations remain largely unrealized. Both because there’s extreme political resistance, and because the idea raises complex questions:
- Who pays: governments, or also complicit companies/organizations?
- Who receives: citizens, or also the state?
- What about countries like Ethiopia that weren’t colonized[1] but are still deeply impoverished?
- Should reparations be a one-time transfer or a recurring payment?
- How do we calculate the right amount?
- Should they replace traditional development aid?
These questions don’t have easy answers, but that doesn’t mean we shouldn’t ask them.
What about just giving people money?
While calls for reparations often get bogged down in political controversy, there's another approach that's been gaining traction: direct cash transfers. No conditions, no strings attached, just money sent straight to people in extreme poverty.
This idea isn’t just theoretical. Organizations like GiveDirectly have been putting it into practice for years, transferring money directly to mobile banking accounts in countries like Kenya, Liberia, Uganda, Rwanda, and Malawi. Recipients can spend it however they like, there’s no micromanagement, just trust.
GiveDirectly runs multiple programs, including:
- Poverty Relief: one-time, unconditional cash transfers (~$1,000) to households in extreme poverty (e.g. Kenya, Malawi, Rwanda).
- Universal Basic Income: long-term daily cash payments, and the world’s largest UBI experiment in Kenya (also programs in Malawi, Mozambique, Liberia).
- Emergency Relief: rapid cash support after disasters (earthquakes, floods, conflicts: e.g. Morocco, Turkey, DRC).
- Climate Relief Fund: cash transfers to communities affected by climate-related disasters (e.g. Uganda, Mozambique, Nigeria).
- Refugee Programs: large cash transfers to refugees to support self-reliance (e.g. Uganda, Kenya, Rwanda).
- U.S. Poverty Relief: unconditional cash aid for low-income Americans (over $270 million distributed since 2017).
While GiveDirectly is the largest, they’re not the only charity that does this. For example, a smaller initiative is Eight, which provides monthly cash transfers over two years to villages in Uganda and the Democratic Republic of Congo.
However, whether it’s for charity or for reparations, giving money directly to the poor does raise some questions: Does it actually work? Won’t the money get wasted? Won’t people become dependent on it?
It’s fair to ask these questions, and researchers have been asking them too. But the data tells a surprisingly optimistic story.
What the Research Shows
According to GiveDirectly and a large body of independent studies, cash transfers generally lead to positive outcomes. People use the money wisely, not wastefully. Corruption is rare, thanks to rigorous auditing, identity checks, and follow-up calls. In fact, around 90% of donated dollars go directly to recipients (with practically everything else going to operational expenses).
Typical spending includes:
- Food and medicine
- School fees
- Durable goods like furniture and livestock
- Home improvements (such as installing a metal roof)
Instead of creating dependency, they can increase independence by funding businesses, education, or healthcare that reduce future need.
Direct cash transfers have been praised by many independent experts. From 2012 to 2022, GiveDirectly was even listed as a “top charity” by GiveWell (a leading charity evaluator) and they still considers cash transfers to be one of the most effective poverty interventions available.
What’s the difference?
These two approaches have many things in common. At their core, they both aim to shift wealth from the Global North to the Global South. And both respect the autonomy and dignity of those receiving support. Whether through justice or generosity, these approaches say: “You know best what to do with this money.”
No conditions. No paternalism. No requirements to attend a workshop or follow a government-approved plan. Just trust. That in itself is powerful. It avoids the “white savior” dynamic that has often plagued development aid and centers the dignity of those who have long been marginalized.
There are also differences. Reparations are, as mentioned, often rooted in compensatory justice, whereas cash transfers are typically justified through a utilitarian lens (the idea that we should do the most good for the most people). The latter is less about guilt or responsibility and more about reducing suffering efficiently; so cash should go to poor people regardless of whether the cause of their poverty was (neo)colonialism.
At first glance, that’s a big divergence. One sounds like justice, the other like charity. But in practice, there’s not much of a difference.
Many decolonial critics argue that almost all extreme poverty today is tied, directly or indirectly, to colonialism or its modern offshoots like structural adjustment, trade imbalances, or climate injustice. Even countries that weren’t colonized, like Ethiopia, have suffered from the ripple effects of imperialism. So while the philosophies may be different, the recipients will be nearly identical.
A Shared Vision
Both approaches, at their best, are motivated by moral outrage. Not pity. They’re responses to a world where billions suffer needlessly while others thrive off inherited advantages.
The desired outcome is the same: an end to extreme poverty, global inequality, and neocolonial control over poor countries' futures.
The biggest difference isn’t the goal, it’s how we get there.
- Reparations would require formal, political action: governments negotiating, apologizing, and transferring funds.
- Cash transfers are mostly handled by NGOs and funded by private donors.
Although even here change is coming. Rory Stewart (Former UK cabinet minister and former CEO of GiveDirectly), has suggested that governments should include direct transfers in official development programs. That would make them only one apology away from becoming reparations.
And of course, there’s no reason we can’t do both. I’d love to see more conversations between advocates of reparations and champions of cash transfers. I think it could lead to very fruitful dialogue and coordination. These approaches aren’t mutually exclusive, they’re potentially complementary.
In the meantime, please consider sending some reparations yourself via GiveDirectly.
A huge thanks to Maxim Vandaele for his tremendous help with this post. All opinions and mistakes are my own.
- ^
Although Ethiopia was very briefly occupied under Mussolini, from 1935 to 1941
Reparations and direct cash transfers seem like totally different things to me. GiveDirectly is about rich people giving to poor people; reparations are about (the descendants of) bad people giving to (the descendants of) their victims. Even if you believe in this sort of inter-generational blood debt and handwave aside the non-identity problem, there is the direction of payment will often diverge from GiveDirectly. For example, many of the descendants of slavers presumably remain in Africa and are quite poor, while the descendants of many of their victims now live in America and have significantly higher incomes.
Bob, do you know of any good sources that go into detail on unfairness and exploitation in the global trade systems? Maybe it's a lot to ask, but ideally I'd want a source that was:
1) Written from a fairly radical, probably therefore socialist, but at any rate, anti-Western capitalist perspective.
2) Actually written to try and persuade someone like me, a fairly economically centrist, broadly pro-market, pro-welfare state social liberal, rather than just assuming a lot of premises that someone like me would disagree with (especially, assuming those premises at crucial points in the argument.)
3) Goes into a fair amount of empirical detail, and is generally scope-sensitive, and able to acknowledge trade-offs (for example, that at least potentially the same development can harm the environment but hurt people, or create both winners and losers in the same developing country.)
4) Is relatively non-technical, so that I can at least get some sense of what it's argument is, despite not having a social science degree.
(Asking partly because my girlfriend is a communist and challenged me to read into this more!)
Maybe nothing meets those requirements (and I'm not saying I'll only read something that does), it's a fairly demanding list. But I thought it was at least worth asking.
Hi David,
I don't like commenting on the EA Forum given the karma-system's distortionary effect, so your chances of getting a response are much higher if you use substack/reddit/DM/email/any-other-medium. However, since you addressed me directly I'm not going to be so impolite as to ignore you, so I'll give it a go.
Unfortunately, I'm not sure I have one that meets all these requirements. I linked Jason Hickel's "The Divide", which is probably the closest, but it's been half a decade since I read it. Given that 1: I have trouble remembering what I even had for breakfast, and 2: I don't even remember if I finished it, I don't think I'm in a good position to recommend it.
I heard that Ha-Joon Chang's "Bad Samaritans" tried something like what you mentioned, but I haven't read it. Writing for an antipodal/hostile audience is quite an unrewarding job (trust me), so I don't know how many writers you'll find. I'll keep my eye out for one though.
I also tend to search literature by topic, and not by author-affiliation, so I most often don't know the political position of the people I'm reading, beyond the vague vibes the text itself is giving off.
I know the World Bank did some dodgy shit, which socialists have raked them over the coals over, but I'm not sure if capitalist scholars (today) defend those actions, so I'm not sure if it's a uniquely socialist critique.
I think, given your profile, you'll probably find the arguments centered on western (farming) subsidies the most convincing (some sources linked in the beginning, though sources more to your liking are likely available online), and you'll probably find the arguments centered on "unequal exchange" the least convincing. (or, well, Open Borders and Climate Injustices are the ones you'll likely find the most convincing, but given that many capitalist scholars also champion those, I think you already believe in those)
But also, I'm not an expert on global trade. The argument for reparations stems more from what happened in history than what's happening today. Some highly upvoted comments by EAs were defending colonialism (which I know a bit about, so I'll probably do a post on that at some point), but for global trade I've only read a small handful of books and articles on the topic, so I'd have to look into it more.
Thanks for responding
I tend to generally agree with a net unjust-enrichment theory of obligation, although I think it is rather difficult to apply in practice.
However, I'd note that "what was taken" and what you have benefitted may not be two sides of the same coin. In fact, I think they often are not, especially given the passage of time.
I'll use an individual example for simplicity and to try to avoid getting into a discussion of any particular injustice rather than the general idea. Suppose my great-grandparents ran whatever the early 20th century version of a big crypto scam is and got away with it. It's likely -- but not certain -- that I have accrued some advantages in life as a result of this bonanza, and that at least some of these advantages [1] would constitute unjust enrichment that I should disgorge. However, the extent to which I (and my living relatives) have benefitted from the scam's existence may be one to several orders of magnitude less than the degree of harm caused.
Indeed, for many significant harms, the correct amount of reparations on an unjust-enrichment theory is ~$0 -- because there is no living person who accrued benefit from the injustice, or the person who did so is sitting penniless in a prison cell for the foreseeable future somewhere. This is a serious practical shortcoming of the criminal restitution system.
I could imagine that some historical and widespread injustices could fall into this category too -- if respect for human rights, an educated workforce, ordered liberty, and so on are rather good for society and for economic development, then it follows that I'd probably be better off if my society had done a better job respecting those things for all members of my society than in the actual world.[2] I don't want to make that assertion about any particular past injustice, certainly not on an unresearched EA Forum comment. But I mention it to illustrate the complexities that arise when one ties the obligation to disgorge unjust enrichment to the payer's personal benefit.
In the end, I suspect that I am a net beneficiary from injustice and that I should view at least some of my charitable contributions as a form of disgorgement (and I do). But my margin of error here is very large.
That raises another question:
Determining which advantages would be tricky, and there could be no expectation of precision here. Suppose, for instance, my parents were able to start a business with some of the scam money (and would not have been able to otherwise), but the business was ultimately successful because of their skill and hard work. Neither deeming the business profits wholly cleaned nor applying but-for causation seems satisfactory here.
Over enough time, this might even be true with respect to other societies -- if prior generations in my society had treated people from other societies more justly, at some point I suspect that I would be better off than I am now because I'd have more prosperous trading partners, would benefit from more technological progress coming out of those countries, and so on.
The compensatory justice view, as you articulate it, suffers from another major theoretical flaw, even contemporaneously: Whether you do it or not seems heavily empirically contingent upon not just the scale of harm, but the scale of benefit you get from it.
Take slavery. A common anti-slavery talking point in the 1800s was that slavery was net harmful not just to the enslaved but to the slavers as well. These arguments always seemed quite plausible to me. The economic benefits of slavery seem dubious, and it seems prima facie likely that the evils of slavery are morally corrosive.
However, if you take that view seriously, you'd also have to take the view that you do not own compensatory justice for slavery. This seems surprising to me, as a matter of ethics. Usually if you do something bad you ought to compensate your victims. When the courts find someone guilty but do not order recompensation, it's usually because the aggressor is unable to pay (As philosophers say, "ought implies can"). I'm not a legal expert, but I do not think it's considered an appropriate defense to say "Your Honor, according to my psychologist stealing has caused me to develop a dependency mindset and I actually have counterfactually lower wealth than if I didn't steal. Thus, I should not pay my victims." And yet this appears to be the compensatory justice view!
I agree with that -- I don't view the unjust-enrichment approach as the only pathway that may require someone to pay amends. The person who actively did something wrong -- say, a drunk driver who crashes into another vehicle -- has an obligation to compensate their victims, and that obligation isn't contingent on the wrongdoer having received a benefit.
If one views the prior wrongdoers as governments, one could attempt to apply this logic to past wrongs, but it would come across to me as visiting the sins of the fathers and mothers onto their children. As a practical matter, governments are ultimately funded by the resources of their citizens and other taxpayers.
Once we move beyond the person who actively committed the wrongdoing, the more direct pathway usually isn't going to work. In my hypothetical, my scamming great-grandparents are dead, so there's no compensation to be had from them. The question is what to do about any passive beneficiaries who were not morally culpable in the commission of the original wrong but have received net benefit from it.
Suppose my great-grandparents realized the benefits of intergenerational wealth, and so I receive a payment from the Scamming Ancestors Family Trust every month. I assert that I am unjustly enriched and incur a moral obligation to disgorge. In this unrealistic toy example, the obligation may extend to the full amount of the payment (although I think this would be uncommon in real-life practice). But it extends no further than that. If my relatives invested the Scamming Ancestors Family Trust in bad crypto and so I never got any payments, I have no moral obligations as a result of its prior existence. The victims are out of luck in that situation.
In the toy example, the existence of a trust makes it fairly easy to see that I personally continue to derive a benefit from the past injustice. In most real-world situations, this will be more difficult to determine.
Yeah after thinking about it somewhat more I'm tentatively skeptical about the case for reparations.
In addition to what Linch said, another complexity here is that compensation can seem plausibly owed to people even in situations where they have not been left worse off by the original wrong. Long ago, Derek Parfit pointed out that if you significantly change the course of the future for the worse, the people who exist in the future will be different than if you hadn't. (The original example was, deplete resources, and the people who exist in the impoverished future would not have exist in the nicer future where you left sufficient resources for future people.) Therefore, as long as those people have lives worth living, the people in the future you spoiled are actually not worse off because of your bad action. They are either overall better off, because they got to have lives worth living, or neither better nor worse off, if you think you can't compare someone's well being in a scenario where they do exist, to their well-being in a scenario where they don't. Nonetheless, it's still plausible that in some cases future people can be owed compensation for a wrong action that led to their worthwhile existence. I can't just say "oh, well, if I hadn't stolen that money from your Dad, he'd never have met your Mum, therefore you wouldn't exist, so I don't need to pay it back to you as his heir". (Why? I think at least because there is value in maintaining a system of compensation for wrongs that goes above and beyond making things intrinsically more fair, but there may be more to it than that and deontologists probably think there is more.)