Doing Good is No Place for Emotion: Podcast with Will Macaskill

Summary

The CGD Podcast: "Doing Good is No Place for Emotion" with Will Macaskill Center for Global Development • August 4, 2015

---

Episode Overview

This episode explores the principles and impact of the effective altruism movement, focusing on rational, evidence-based approaches to doing good in global development. Host Rajesh Merchandani interviews Will Macaskill, co-founder of Giving What We Can and author of Doing Good Better. Together, they discuss the pitfalls of emotionally-driven charity, the need for rigorous evaluation of aid interventions, and how effective altruism is shifting the landscape of giving and development strategy worldwide.

---

Key Discussion Points & Insights

1. Defining Effective Altruism

• Effective altruism is about making the most positive impact possible with your resources—using "good evidence, careful reasoning" to ensure you "make the most difference we can." (B: 00:42)
• Distinction drawn between standard development practices and the more rigorous reflection called for by effective altruists.

2. Learning from Failure: The PlayPump Example

• The PlayPump (a merry-go-round that also pumped water) is cited as an example where "sexiness" won over practicality. Despite millions in support, it proved costly, inefficient, and ultimately unwanted by communities. (B: 01:13–02:42)
• Quote: "This was a great example of sexiness over practicability. It was a disaster from the start, really." (B: 01:44)

3. Evidence-first Success: The Deworming Case Study

• Contrasted with PlayPump, Michael Kremer's research used randomized trials to identify genuinely effective interventions.
• Providing textbooks showed no impact, but mass deworming with albendazole (costing ~50 cents per child) significantly improved school attendance and long-term economic outcomes. (B: 03:50–04:31)
• Quote: "...willingness to challenge your assumptions...be willing to experiment...what’s actually going to have the biggest impact, even if that's something as unsexy as deworming." (B: 04:34)

4. The Growth of the Effective Altruism Movement

• Organizations such as Giving What We Can and 80,000 Hours emerged to help people maximize the impact of their giving and career choices.
  • Giving What We Can: Members pledge ≥10% of their income.
  • 80,000 Hours: Guides individuals on impactful career decisions.
• The movement is expanding, with hundreds of organizations adopting the approach and millions of dollars pledged to evidence-based charities.
• Quote: "You can maybe increase the amount you donate by tenfold... you can increase how effective your donations are 100 fold." (B: 05:10)

5. Emotion vs. Rationality in Doing Good

• Emotional motivations to help are acknowledged as natural but must be guided by reason to ensure maximum impact.
• Dilemma: Personal losses motivate giving to certain causes (e.g., cancer), but effective altruists urge channeling that towards where the need is greatest—even if less personally salient.
• Quote: "It is a little bit more abstract than the way we normally approach charity, but I think it's just as human." (B: 07:58)
• The appeal of disaster relief is largely emotive, but ongoing, underfunded crises (malaria, TB) may present greater opportunities for impact.
• Quote: "We need to more rationally reflect in order to think, well, is this the best way I could be doing good?" (B: 10:26)

6. Donation Priorities: Immediate Relief vs. Systemic Change

• Macaskill advocates for donating to highly effective, measurable interventions, e.g.:
  • Against Malaria Foundation (bed nets)
  • Deworming charities
  • GiveDirectly (unconditional cash transfers)
• Acknowledges the importance of policy change for large-scale, long-term impact.
  • CGD is praised as a leading organization in this arena.
  • Quote: "Among organizations trying to affect policy change...I don't know of any organization better than CGD." (B: 11:49)

7. Effective Altruism & Existential Risks

• The movement is expanding its scope beyond global poverty to existential threats:
  • Artificial intelligence, pandemics, biosecurity, geoengineering
• Emphasizes the "three-factor framework":
  • Scale of the problem
  • Neglect (how overlooked it is)
  • Tractability (how solvable it is)
• These areas are often underfunded by governments, so philanthropy can have outsized impact.
• Quote: "If you think ideally the states would be funding this, but at the moment they're not. Well, maybe I can create a real societal shift there with comparatively small investments." (B: 16:08)

8. The Future of Aid and Development

• Direct cash transfers are likely to play a far larger role; interventions should be assessed against the baseline of simply giving cash.
• Emphasis on more research and randomized trials to determine what works; more spending to optimize health outcomes globally.
  • Health interventions often have broader applicability ("external validity") than educational or institutional reforms.
• Quote: "If you think your program is not going to do more good than just giving people cash, then what on earth are you doing?" (B: 17:53)
• Aid should focus more on global health, direct giving, and rigorous evaluation of interventions.

---

Memorable Quotes & Timestamps

• On "sexy" ineffective interventions:
  "This was a great example of sexiness over practicability. It was a disaster from the start, really." (Macaskill, 01:44)
• On identifying effective interventions:
  "...be willing to experiment if that's possible, but willing to think of the different things I can do, what's actually going to have the biggest impact, even if that's something as unsexy as deworming." (Macaskill, 04:34)
• On maximizing donation impact:
  "You can maybe increase the amount you donate by tenfold... you can increase how effective your donations are 100 fold." (Macaskill, 05:10)
• On emotional vs. rational giving:
  "...care about preventing natural disasters...just today, 70 jumbo jets worth full of children died as a result of easily preventable diseases... that’s not covered in the media. But obviously it's just as big a catastrophe as natural disasters." (Macaskill, 10:42)
• On the role of policy:
  "I certainly think that there's enormous potential to do a huge amount of good through policy change." (Macaskill, 11:49)
• On cash transfers as a baseline:
  "If you think your program is not going to do more good than just giving people cash, then what on earth are you doing?" (Macaskill, 17:53)

---

Important Segments (Timestamps)

• [00:42] – What is effective altruism?
• [01:13–02:42] – PlayPump failure as cautionary tale
• [03:50–04:31] – Deworming as effective intervention
• [05:10] – Growth and impact of effective altruism movement
• [07:58–10:26] – Emotional motivations vs. rational impact in charity
• [11:36–11:49] – Policy change and the role of organizations like CGD
• [13:20–16:08] – Addressing existential risks (AI, pandemics) through effective altruism
• [17:53–18:38] – Future of aid: cash transfers, health, and rigorous evaluation

---

Final Thoughts

Will Macaskill advocates for a shift in the way we approach global development and philanthropy—moving from good intentions and emotional responses to a rigorous, evidence-based focus on outcomes. His vision for the future of aid is rooted in maximizing impact: more empirical research, higher standards for effectiveness, prioritizing global health, direct cash transfers, and rational consideration of neglected, high-stakes risks. The episode serves as both an introduction and a rallying cry for effective altruism’s growing movement.

---

Transcript

[A]

Hello, I'm Rajesh Merchandani and welcome to this edition of the CGD podcast. The question today is, how do we do the best good? That's the subject of the Effective Altruism Global conference at Google's campus in California, where I've come to speak with one of the leading lights of the effective altruism movement, Will MacAskill, whose new book, Doing Good Better, gives us an idea of what we should and should not be investing in if we want to make a positive change in the world. Will, hello. Great to talk to you.

[B]

Great to speak to you.

[A]

So, for those who don't know, explain effective altruism.

[B]

Effective altruism is about using your time and money as effectively as possible to make the world a better place. It's about using good evidence, careful reasoning, attempt to make not just a difference, but to make the most difference we can.

[A]

Now, a lot of people would say, working in development, they would say, well, that's what we do already. We are careful in the selection of what we invest in and careful in the projects that we actually set up. But that's not enough for you, is it?

[B]

No. I want to be ensuring that we have the biggest impact possible. And I think far often people are too unreflective when they try to do good. So in my book, I tell the story of the play pump. This is innovation that got a huge amount of attention in the noughties. It is a children's merry go round that would also function as a water pump. So children would push this merry go round and play on it, and that would pump clean water.

[A]

And it seemed to be like a big new thing.

[B]

It seemed very exciting. It got millions of dollars of investment that USAD Aid and the Case Foundation. Jay Z went a tour to promote it and people were very excited. Bill Clinton promoted it. It won a World Bank Development Media Marketplace award. And so people were very excited about this. But this was a great example of sexiness over practability. It was a disaster from the start, really. It cost three times as much as a traditional hand pump, yet would pump less water. It required constant torque to keep spinning, unlike a typical hand pump, a typical merry go round. So children would get tired very quickly and they wouldn't want to play on the pump. So it was left up to the women of the village to push the pump, a task they found undignified and demeaning. What's more, the communities had just never been asked, do you even want this? If they had been asked, they would have said they Preferred the hand pumps that they had in particular wanted playground immunity. They wanted easy to pump clean water.

[A]

Okay, so that would be an example of what you would call ineffective.

[B]

That's exactly how cool.

[A]

So what should they have done?

[B]

So instead, I mean, I think story I contrast them with is Michael Kramer going and working in Kenya and then starting to working with a charity called ICS who is doing a variety of educational programs. And then thinking, okay, let's actually test these. Let's look at, take seven schools and just, just monitor them, not do any intervention. That's the control group. Take another seven schools and deploy our intervention that might be providing textbooks or providing flip charts or more teachers. And then see, well, how effective is this program? Is it better than doing nothing? And what they actually discovered was that very often an idea or program that seemed like it would obviously benefit people, for example, distributing textbooks at the time a classroom would only have one textbook. Obviously you'd think if you increase the number of textbooks available to children, educational outcomes would improve. But they put it to the test and actually discovered that there was no effect at all.

[A]

Yeah. And the thing they found worked was.

[B]

The thing they found worked was deworming school children. Something you might not have even thought of. So albendazole, a drug developed in the 50s and now off patent, it can be distributed en masse to school children costing only about 50 cents per child. And that was discovered to be one of the most cost effective ways of improving school attendance. They found that children weren't attending school in part just because they were sick. They had worms living in their guts that was making them worse off. And then when there were long run follow ups, the children who were dewormed had better economic productivity. They were earning more, they were working more hours, and that was a decade later.

[A]

And so for you, that is a great example of effective altruism.

[B]

That's a great example of effective altruism. A willingness to challenge your assumptions and to actually think, I'm not going to just presume that this thing that seems good actually is good. I'm going to look at the evidence, be willing to experiment if that's possible, but willing to think of the different things I can do, what's actually going to have the biggest impact, even if that's something as unsexy as deworming.

[A]

Now, you're not the only person in the effective altruism space. Obviously there are quite a lot of organizations there kind of overlapping a bit. Now you know them all, you're all mates and most of Them are here in San Jose at this conference. And there's a real sense of a movement growing.

[B]

That's exactly right. So I co founded an organization called Giving what we can in 2009, and that had two aims, trying to get people to give more. So every member of Giving what We can pledges at least 10% of their earnings and some go much further. So myself and Toby Ord, the other co founder, we've both committed to give most of our earnings over the course of our lives, everything we earn above 20,000 pounds. But then the second aspect is encouraging people to give more effectively. And when we launched, we made some recommended charities, including Deworming Charities. And that's actually the even bigger aspect because you can maybe increase the amount you donate by tenfold. Let's say you can increase how effective your donations are 100 fold. And out of that came an awful lot of interest from people who realized, yeah, finally I want to actually know that I'm making the biggest impact that I can. A lot of people came to me asking, okay, you've done this great work on what charities are most effective, but I have this big career decision I need to make. What should I do then? And partly in response to that, I started doing research onto that question and launched an organization called 80,000 Hours, which gives, obviously it's much harder to quantify many of this, but tries to give as rigorous and reflective an analysis as possible of what careers can do the most good in other ways. The movement has started to explode. So the term effective altruism came into being in 2012, and people realized this is a very important concept. Doing more good, doing it more effectively. And now there are hundreds of people setting up organizations or running organizations based around these ideas. Thousands of people around the world who would self identify as effective altruists. And we're already achieving amazing things. Giving what We can now has 1,000 members with $450 million of lifetime pledge donations. GiveWell, which does outstanding in depth charity recommendations, moved $28 million last year. And Good Ventures, a foundation it advised, donated $8 million in grants to other causes. So there's definitely a feeling of a big movement here. This is the start of something very big.

[A]

The numbers are growing. That's interesting to see. It's still a drop in the ocean. It's going to take a lot to kind of persuade the world to adopt effective altruism. And for this reason, maybe that, you know, if you say to someone, okay, you had a relative that died of cancer, but you should not go and raise money for cancer charities. You should go and give money to end malaria because it's more effective in a way that takes the personal connection out of altruism. For many people, it almost takes the humanity out of humanity.

[B]

Yeah. So I think that's a reasonable concern. And I think that the impulse to want to do good, you know, if a family member dies, for example, is obviously a natural and an admirable one. But the way I think of it is, so I have family members who died of cancer, but I think supposing they died of something else, would that have been any less tragic? Of course not. What I ultimately cared about there, what gives me the moral motivation, is that someone died and suffered before their time. Whether it was from one disease or another, is neither here nor there. And I think that's therefore what we should focus on. If someone close to you dies or suffers, you should take that moral motivation, but turn it to just reducing suffering and reducing early death in general. And so it is a little bit more abstract than the way we normally approach charity, but I think it's just as human.

[A]

That's interesting because you could argue that actually it's just being entirely rational about altruism. So it's kind of machine like.

[B]

Well, I mean, I think that we should be more rational, should be more reason based when it comes to doing good. Because I think if we're unreflective about where we're giving, or if we just give on the basis of emotional salience, then we're just going to do less to make the world a better place. And I think that's in conflict with maybe our deep values in particular. If you just focus on the things that are most emotionally salient, then you're probably doing fairly similar activities that everyone else is doing. And that means you're probably focusing on something that's already comparatively well funded. And so I talk about this in my book with respect to disaster relief versus what I would call ongoing disasters like tuberculosis, malaria, tropical diseases, natural disasters, obviously we should fund them. Obviously not fund natural disasters, but fund disaster relief. Disaster relief, obviously we should fund that. But they get far more in funding comparatively than these ongoing natural disasters because they're so much more emotionally salient. And so I do think that the correct way to approach doing good is just to have this combination. You have the natural human emotional impulse to help, but then that does need to be guided by our ability to reason and to think. That means that our first instincts might not be the correct ones. We need to more rationally Reflect in order to think, well, is this the best way I could be doing good?

[A]

So when you see the pictures on the TV of a natural disaster, you know, it tugs on your heartstrings. You pick up the phone, you donate $100, 100 pounds. Whatever you're saying, don't do that. Take a moment, think more rationally, do some research about how that hundred dollars, one hundred pounds could be most effective.

[B]

Yeah, that's right. Rather than you think this natural disaster, that's terrible, well then think there's ongoing natural disasters. Just today, 70 jumbo jets worth full of children died as a result of easily preventable diseases. And that's not covered in the media. But obviously it's just as big a catastrophe as natural disasters. And if we care about preventing natural disasters, well, and preventing the scale of suffering caused by them, making countries economically better off is a great way to do that. And the best ways to do that, that I know of, at least among concrete, measurable ways of doing so are things like donating to against Malaria foundation that distributes insecticide treated bed nets. Donating to deworming, or donating to give directly which just simply transfers cash to the poorest people in the world.

[A]

Or if you want to help developing countries create stable, prosperous, well governed economies, you can tackle policy change as CGD does.

[B]

That's right.

[A]

And we've come out pretty well in your assessment of altruism.

[B]

That's right. So when I said that, I was careful to say concrete, measurable ways of doing good. This is a big issue for the effect of altruism community and for people trying to do good is how much do you put weight on things that you know will definitely have some impact versus things that are harder to quantify, maybe a bit more speculative, but potentially having a very large impact if they are successful. I think if anyone says it's obvious what you should do either way, then they're being overconfident. But I certainly think that there's enormous potential to do a huge amount of good through policy change. And among organizations trying to affect policy change in ways that will benefit the extreme poor. I don't know of any organization better than cgd. Get away now.

[A]

Okay, we'll get back to the point. Thank you very much for that. I'll get a tick for that. Okay, you deserve it. You talked a bit earlier about thinking rationally, almost being a machine at the conference, one of the big focuses of many of the delegates here. And you know, remember where we are, we are in Silicon Valley. There are a lot of very bright entrepreneurial tech types. There's a lot of people talking about artificial intelligence. The risks of artificial intelligence, killer robots to you and me, or not you to me. And problems like existential risks caused by things like pandemics. Are these suitable topics for effective altruists?

[B]

I think they certainly are. So the primary focus of most of the movement is on extreme poverty, but that's just one cause. There's many social problems in the world and the way we think about what causes to focus on normally uses this three factor framework. So one is what's the scale of the program, the problem, how many people are affected, what amount of value is at stake? Second is how neglected is this problem? Is it the case that already huge numbers of people and a huge amount of resources are being focused on it and so the difference you'll make in addition won't be so great? And then third is how tractable is this problem? How much progress can you make on it with a given amount of effort? In the case of risks from new technology, so small probability risks of very large damages, the scale is very great. We're worried about, sure, the small probabilities, but small probabilities of things that could end civilization. So it's very great in value. Also, it seems very neglected, especially artificial intelligence, because it seems like the sci fi thing. It's still quite far in the future. So at least until recently and even now it's quite neglected, a very neglected area.

[A]

Actually. Elon Musk was on a panel today and he was saying it's closer than we think. Artificial superintelligence.

[B]

That's right. There's been a lot of progress in artificial intelligence over the last five years. It's still not on our doorstep. I mean, I think a reasonable estimate is that we'll develop human level intelligence at least approximately sometime in the next 20 to 70 years. I think that's more likely than not. Obviously some chance we'll just never do it. But if you are survey experts, that's what they think.

[A]

But is this an area for private altruism? This is what states should be investing in it, isn't it, in research into these areas? This is a global public good.

[B]

Yeah. I think ultimately states should be investing in it. Currently they're not. And one thing philanthropy can do is for these sorts of issues. So what's the incentive of a government to start focusing on this? It's not going to affect them in the four year, five year election cycle. And so one thing philanthropy can do is get a Certain cause on the agenda. And also, I mean at the moment with something like this, it's still in the future. It's just something we want to be have foresight about. We need to be doing policy research. Like at the moment I can't think of any AI related policies that I would happily stand behind. And so that means to begin with, we're just at the stage of just figuring out how important is this and how much work should we be, how much work should we be doing on this and what sort of policy for the EMTIF policy. The sponsors are suitable, but there is.

[A]

Vastly research being done in the area of things like pandemics, existential risks.

[B]

Yeah, that's right, there are, but there.

[A]

Should be more of that.

[B]

I think there should be more of that.

[A]

Private philanthropists are not going to pay for that.

[B]

I think private philanthropists, I mean, I think in the ideal world perhaps states would fully fund this. I think we're not in the ideal world. And so there's very little research. There's quite a lot on pandemics, especially, especially from security. Whether there are gaps that might be filled by philanthropy is another question. Much less work on other sorts of existential risks. Very little work done on geoengineering. I think that a lot of more people should be thinking about that as a technology that just will be developed and perhaps predictably will be deployed. We should be really concerned to be investing in that. And then these things that see more science fictiony, such as artificial intelligence, certainly aren't being funded at the moment. And I think it can be a great opportunity for private philanthropists. If you think ideally the states would be funding this, but at the moment they're not. Well, maybe I can create a real societal shift there with comparatively small investments. I helped to create a small think tank called the Global Priorities Project which does do the search on this. On the policy side, the Future of Humanity Institute, center for Study of Existential Risks are university based institutes that are also working on these questions.

[A]

Let's get back to kind of more of CGD's comfort zone, global poverty rather than artificial intelligence. Although I know that there's a lot of real intelligence amongst my experts. Mine's just the artificial stuff. In the area of global poverty reduction, look, 10 years, 20 years down the line, say more people get involved in effective altruism. What is the landscape of development, the landscape of aid then if people adopt your principles?

[B]

Yeah, I think that's a great question. The thing that leaps out to me is something that I think will grow a lot is direct cash transfers. I think it's remarkable that this isn't to give directly, which just transfers cash to the very poorest people. It's seen as a very innovative delivery, development, charity. And from one perspective, that's just bizarre. It should be the default. If you think your program is not going to do more good than just giving people cash, then what on earth are you doing? And I think that that's a very useful standard. We can use this as a yardstick because this is a simple, very easily scalable, very low overhead thing that you can do that we have good confidence works, but for both theoretical and empirical reasons.

[A]

Do you see a future where a lot of aid agencies don't exist as.

[B]

They do at the moment and instead just transfer cash or whatever you think?

[A]

I mean, oh yeah, maybe direct cash transfers, using something like kid directly doing more policy research, but just a difference. Because you don't think people should be giving to pet causes.

[B]

Yeah, I mean, I think the landscape could be very different. I think there could be a lot more, far more research done to work out what works. I'd love to see a much larger share of spending going on. Trials and assessment of what works in general, I guess I think more should be spent on global health relative to other things. That's for two reasons. One is because global health has seen such amazing success stories in the past, such as the eradication of smallpox and progress on other immunizable diseases. And then secondly, because of what economists call external validity. So if you know that a deworming drug will kill worms of a child in one country, that's going to work in many other countries as well. Because human bodies are pretty similar all around the world. In contrast, if you have an education program, some new education program that works very well in one country, Education systems differ quite a lot. You don't know if that's going to be as effective elsewhere. And then finally, through health, you're giving people something that basically everyone wants. Rawls called this one of the bases of self respect. You know, you're not going to be interfering particularly or decreasing the risk of interfering with someone else's culture by say, changing their education system. Because basically everyone wants health. Having good health is a pretty requisite for doing many other things. So I think the arguments for global health are generally stronger than for other sorts of development. So I think the three things of cash as a baseline, more investment in health and much more investment in actually figuring out what's effective and just doing what's effective. Are the three things I'd like to see a lot more of.

[A]

You talked about cash transfers. CGD's got some work on humanitarian cash transfers coming out in a few weeks time. Not going to give you the scoop on that just yet. You have to wait and see for now. Will MacAskill, fascinating to talk to you. Thank you very much for joining us on the CGD podcast.

[B]

Thank you so much.

[A]

You can find out much more information about everything that CGD is doing. As [e-mail] I'm Rajesh Merchandani and join me again for the next edition of the CGD podcast from the center for Global Development.