Johan Andersson

Software Engineer with an M.S. in Computer Science. Based in Lund

Notice: migration to in progress

Distilled notes on "Doing Good Better"

13 Apr 2018

When trying to improve the world, one should not solely focus on doing. One should also focus on doing where it matters most. This is the main theme in William MacAskill’s book Doing Good Better.

This blog post is not an attempt at introducing the concept of Effective altruism (EA), but rather an attempt at summarizing the contents of Doing Good Better. Summarizing this info will surely give a newcomer some sense of what EA is, but if you’re new to this area there are better ways to get introduced.

The primary goal of this write-up is to strengthen the knowledge in people already familiar with the basic concept, but who feel they lack content. It is thus not aimed to be some sort of introduction to EA, of which there are already a plenty (including this book).

In short, this is for those who can’t find the time, but naturally it is hard to accurately comprise a 200-page book into one blog post.

Good intentions do not necessitate good effects

In 1995 the PlayPump had its first installment. It was a water pump designed for kids who would spin it around and have fun, while simultaneously pumping up water from the ground. It was a good intention, that would later be shown to have negative effects.

The project received endorsements from numerous influential people:

  • Steve Case (then CEO of AOL)
  • Jay-Z (Hip-hop artist)
  • Bill Clinton (former president of the US.)
  • Laura Bush (then first lady of the US.)

In 2008, PlayPumps’ founder said:

“It really rocks me to know we’re making a difference to a lot of people who are nowhere near as privileged as I am or my family is”

By 2009, eight hundred PlayPumps had been installed across a couple of southern African countries. It was then discovered that the PlayPump was not so fun after all. While most merry-go-rounds have sufficiently low friction to spin freely after some applied force, the PlayPump needed constant force in order to pump water, exhausting the playing children.

“Much of the time, women of the village ended up pushing the merry-go-round themselves—a task they found tiring, undignified, and demeaning.”

A woman in Mozambique said, “From five AM., we are in the fields, working for six hours. Then we come to this pump and have to turn it. From this, your arms start to hurt. The old hand pump was much easier.”

One reporter estimated that, in order to provide a typical village’s water needs, the merry-go-round would have to spin for twenty-seven hours per day.

Not only did the pumps prove a hurtful replacement to the old hand pumps, but it turned out that no one had even asked some of the affected communities whether they wanted a PlayPump in the first place.

And to top it all of, the PlayPump:

  • often broke down in a couple of months
  • was difficult to repair (and getting maintenance support was difficult)
  • cost four times as much as a hand pump.

Thus, even while the idea of the PlayPump had good intentions, it turned out to be a net-negative experience for the affected parties.

Watch the Frontline/World episode ‘Troubled Water’ (2010) describing the issues of the PlayPump here. Start at the 3 minute mark to skip the intro. Note the good intentions.

The lesson

Don’t evaluate a charity by judging its intentions; evaluate it by judging its effects.

How much can you do?

Even though your country may be divided in its wealth distribution and you may feel poor in comparison to the richest of the rich, you are in all likelihood in the top from a global perspective.

  • income above $52K/year: top 1% globally
  • income above $28K/year: top 5% globally
  • income above $11K/year: top 15% globally

A student in Sweden with maxed student loans and allowances has a yearly income of about 110,000-120,000 Swedish kronor, which is roughly $13K-14K US dollars. This puts Swedish students in the global top 15%, even though Swedish students are commonly referred to as relatively poor within the country.

Chances are that you, the reader, are either a student with a comparable income to the Swedish student or someone in the working world with a greater income. Even though you might feel like you’re not well off in your own country, you can still make a considerable difference to people much poorer and worse off than you. And if you feel like you’re well off, why not try to spread that feeling to others in the world?

You are able to do something. But what should you do?

Weighing suffering

Arguably, one of the best things you can do in the world is to reduce others’ suffering. But rather than just arbitrarily choosing some form of suffering, should we not try to optimize the amount of suffering we can reduce? If so, we would need a method of comparison.

It turns out however that it is hard to compare different sorts of suffering. How would you compare the suffering of someone with cancer and someone with AIDS? To help, we can try to quantize suffering and use a metric for comparison, in order to evaluate where the need for support is the greatest.

One such metric is QUALY (quality-adjusted life years). Using a metric like QUALY helps us in deciding who we should help.

For example: should we donate $4000 to a guide dog training charity in order to train one guide dog for one person suffering from blindness, or should we donate $4000 to a charity fighting trachoma and cure hundreds of people suffering from blindness?

In which scenario do we gain the most QUALYs for our buck? Obviously, the latter.

If you don’t want to fight on the front lines

You don’t have to join Doctors Without Borders. You don’t have to volunteer in poor, remote villages. You don’t have to give up your privileged urban life in order to help people.

Instead, you can save people and make a real difference by donating some of your income to effective charities (i.e. not PlayPump). If you’re a student, try donating 1% of your income. If you’re working, see if you can donate somewhere up to 10% of your income.

The cost of saving a life in the developing world is about $3400 (this pays for 560 mosquito nets which on average saves 1 person from dying in malaria). If you earn $28K a year and donate 10% of your earnings, you would save a life every 440 days. In your lifetime of, say, 40 working years you would then save approximately 33 lives. That’s amazing! (but remember! you have to donate effectively for this estimate to hold)

Why you shouldn’t donate to disaster relief

Consider two “recent” disasters:

  • Earthquake in Fukushima, Japan (2011)

    • Millions of people left without electricity or water
    • 15,000 dead
  • Earthquake in Haiti (2010)

    • 280.000 buildings collapsed
    • Cholera broke out
    • 150,000 died

In both cases there was massive global media coverage and large humanitarian relief efforts. In each case the international aid raised amounted to about $5 billion.

Even though Japan is thirty times richer than Haiti (per capita) and even though 150,000 people died in Haiti (vs 15,000 in Japan), the raised international aid resulted in roughly the same amount.

If the global community had been rational we would not have given Japan as much as we had given Haiti, it being a poorer country and much more affected.

It seems that donors don’t consider differences in scale and severity of disasters. Instead it seems that people donate to widely publicized disasters. (A 2008 earthquake in China that killed 87,000 people only received $500 million in donations - probably due to its lesser media coverage)

Funding seems to be allocated in proportion with how evocative and widely publicized the disaster is, rather than on the basis of its scale and severity.

If you feel despondent by being told to not donate to people in immediate danger, consider that there are other people in immediate danger too - only they are not covered on the news. Consider supporting those people instead. GiveWell curates a list of effective charities where your donations actually make a good difference.

What if you chose a career that makes a lot of money

See 80,000 Hours’ piece on this.

TL;DR: find a high paying job you like in order to give and help more.

You might also find this one interesting.

Confusions in regards to charity evaluation

You should not care about how much the CEO is payed or how much overhead a charity has. These are redundant variables that can trick you in evaluating effectiveness. The most important thing you need to ask is: how much good does this charity accomplish for every dollar I give it?

MacAskill lists 5 questions he thinks any donor should ask before deciding where to donate:

  • What does this charity do?
  • How cost-effective is the charity?
  • How robust is the supporting evidence?
  • How well is each program implemented?
  • Does the charity need additional funds?

Which causes are most important?

In order to assess which area we should consider, we should try to prioritize causes which have as many of the following characteristics as possible:

  • The problem is of large magnitude (a large number of people are affected)
  • The cause is neglected (its resources are low in comparison to the problem scale)
  • The cause is tractable (can we do anything about it, effectively?)

If you want to get personally involved, you should also consider the personal fit. Given your expertise, connections, and passions, how likely are you to make a large difference in the area?

MacAskill summarized some problems and their traits in a table (I think these are subjective ratings made by MacAskill):

  Scale Neglectedness Tractability
Extreme poverty 3 2 4
US Criminal Justice Reform 1 2 4
International Labor Mobility 3 2 1
Factory Farming 1-3* 4 3
2-4ºC Climate Change 2 1 2
Catastrophic Climate Change 2-4* 2 2
Other Global Catastrophic Risks 2-4* 3 2

* : Depends on your value judgements

1 : not very / least

4 : best / most

What should you do now?

A donation of $3,400 can provide bed nets that will save someone’s life, deworm seven thousand children, or double the income of fifteen people for a year.

You don’t have to be trapped in war in order to save a large number of lives, like Oskar Schindler. You can be like him even in peace time.

Every one of us has the power to save dozens or hundreds of lives, or to significantly improve the welfare of thousands of people. We might not get books or films written about us, but we can each do an astonishing amount of good, just as Schindler did.

MacAskill recommends four main points that you can do in order to make a difference:

  1. Establish a habit of regular (effective) giving. Remember that committing yourself to payments or savings in beforehand increases your chances for success.
  2. Write down a plan of how you are to incorporate effective altruism into your life.
  3. Join the effective altruism community
  4. Tell others about the idea of Effective altruism or giving effectively.

You don’t have to follow those four points, but if you feel empowered by the idea that you can make a big positive change in other people’s lives then consider doing at least one or two of those points.

If you want to go further than these actions, you might wish to take Giving What We Can’s pledge to donate 10 percent of your income. You could read the career advice on 80,000 Hours, or apply for one-on-one career coaching there. Or you might wish to set up a local meet-up group, starting discussions about effective altruism in your area, with your friends or through your church or university. Further information on all these things is available at

Additional resources you might find fulfilling

Acknowledgements, sources, et al

Unattributed quotes are cited directly from the book (version 1).

While I do provide some URLs to external sources, these for are additional reading that I think are relevant. Doing Good Better has been the sole source used while writing this.

Do not count on this to be an accurate summary if you wish to feel that you’ve read the book, since I have probably:

  • cherry picked content.
  • injected my own thoughts/writings.

Instead, if this peaked your interest, consider reading it (I have a physical copy you can borrow if you are geographically near me).

If you find something incorrect with this post, please contact me by mail or submit an issue on GitHub.