Comparing charities: How big is the difference?

Your donations can do an astonishing amount of good. However, the impact can vary wildly depending on where you donate.

The best charities can be at least ten times better than a typical charity within the same area, hundreds of times better than poor-performing charities, and the worst charities can do harm.

Intervention cost-effectiveness in global health in order of DALY per $1,000 on the y-axis, from the DCP2. Compiled from The Moral Imperative Towards Cost Effectiveness by Toby Ord.

This page contains some out-of-date information. We hope to update it as soon as we can, and still believe that the core argument is correct. For a more recent analysis of this topic, see 80,000 Hours' article: "How much do social problems differ in their effectiveness? A collection of all the studies we could find."

Comparing charities

Imagine you had $100 to spend to help improve school attendance of school children in low-income countries. How many additional years of school could that buy?

Providing merit scholarships for girls would result in about a month or two of school attendance (0.15 years). That would seem like a pretty good deal, right? However, if you spent that $100 on school-based deworming treatments it would result in about 14 years of school – that’s almost one hundred times more schooling.

Furthermore, that same deworming program could give an extra year of healthy life for roughly $28-$70 (according to charity evaluator GiveWell). In comparison, new cancer drugs are generally recommended in Australia if their cost per year of healthy life saved is around $45,000-$75,000. A factor of almost one thousand.

At least merit scholarships and new cancer drugs have positive effects – they still improve schooling and save lives. That isn’t always the case. Suppose you were to spend that same $100 on trying to prevent juvenile offending using the “Scared Straight” program. In that case, it’s estimated that would have a negative effect, costing society $29,300 for that $100 invested.

We’ve collated a list of examples at the bottom of this page, but first...

Is this surprising?

Most people find this surprising, but it probably shouldn’t be. We’re used to seeing uneven distributions in all kinds of fields:

• The most profitable businesses are many many times more profitable than the average business.
• A bestselling author far outsells the average author.
• Many investments lose money while some return 1,000 times the initial investment

Furthermore, charities don’t have the same competitive dynamics as the private sector because it isn’t the beneficiary that pays for the intervention. If one company is charging $10,000 for a laptop and another company is charging $1,000 for a better laptop, the first company wouldn’t survive long. However, a donor will often donate the same amount regardless of the impact.

Does this matter?

Yes, it has a real tangible cost. We just notice it less when it’s affecting others (especially if they’re far away in distance or time, or otherwise different enough from us).

When reading numbers that affect others the only difference between 1 and 100 is two little zeros – it doesn’t feel significant. Our brains don’t really intuitively have an emotional sense of scale (psychologists call this phenomenon scope insensitivity).

To get a sense of scale it can help to try and picture the impact very personally.

Take a moment to slowly read and imagine each of these examples:

• You need a life-saving surgery that costs $500,000. Then you find there is another procedure that’s just as effective for only $5,000.
• Your partner is diagnosed with a disease and they only have an 8% chance of surviving with the standard procedure. Then you are told there’s an alternate procedure for the same cost that increases their chances of living to 80%.
• Your beloved family pet is diagnosed with a disease and is only expected to live for 6 months with standard treatment. You find out that you can have an alternate treatment that will give them 6 years of excellent life.
• Your entire family is stranded with a bushfire raging towards you and only have a small motorbike to escape. Then a person driving an empty minibus comes by to rescue you.

Notice that initial dropping of the stomach, followed by an amazing sense of relief? That is what 10x-100x feels like.

The good news is that:

1. Outstanding giving opportunities can be found; and
2. Many of us are fortunate enough to have significant resources to put to good use (most people reading this would be on the global rich list).

A typical American who donated 10% of their income to an effective charity could choose to save an estimated 40 lives over their career (e.g. ~45 years, ~$50k income, ~$5,500 per life saved donating to Against Malaria Foundation according to GiveWell’s estimate).

It’s amazing how we can significantly improve the lives of others if we use our resources effectively.

What can we do?

Charity evaluation is a difficult task for most donors to do on their own, so we’ve put together our giving recommendations to help you get started.

Donate to effective charities

If you’re convinced that it’s important to improve the lives of others, consider taking a pledge to donate a meaningful portion of your income to help improve the lives of others. It can help you live up to your values, meet like-minded people, and inspire others to follow suit.

Take a giving pledge

If you’re driven to have an impact, you may also be able to significantly help others by pursuing a high-impact career, volunteering, or advocating for effective ways of improving the world.

More effective ways to help others

Charity cost-effectiveness comparisons

We’ve collated the following table of examples which illustrate this underlying point by drawing comparisons with publicly available data.

However, there are some things worth noting:

• The best giving opportunities are often hard to precisely quantify.
• You can find much larger gains when comparing a much wider set of options; e.g.
  • Instead of treating similar conditions in a more effective way, you could treat a different condition;
  • Instead of focusing on the wellbeing of people alive today, you could focus on the wellbeing of future generations or animals.
• These numbers are estimates from a range of sources and times, and use varying statistical methods.
• We don’t necessarily recommend all of the charities used in the ‘more effective’ examples below.
• These estimates generally look at the charities average cost-effectiveness (the ratio between all the benefits they provide divided by all their costs), which is likely to differ from the marginal cost-effectiveness (the ratio between all the benefits gained from an additional donation, divided by the size of that donation). We think donors should primarily be interested in marginal cost-effectiveness.

Examples of charities that do harm (negative cost-effectiveness)

• The Scared Straight programme is estimated to have cost society ~$293 per $1 invested (the intervention has increased juvenile offending)
• Play Pumps each cost ~$14,000 and are reported to have been much worse than the cheaper hand pumps they replaced

What are the best and worst charities to donate to?

The best charities are ones that are evaluated to be highly impactful — they work on an important problem and do the most good with the resources they have. The worst charities are ones that actively harm those whom they intend to help, or society at large.

Donate to effective charities

Join our effective giving community

If you've made it this far, we hope you're inspired to give more, and to give more effectively.

Join the Giving What We Can community by taking a pledge to donate a meaningful portion of your income to help improve the lives of others. It can help you to live up to your values, meet like-minded people, and inspire others to follow suit.

Take a pledge

Not ready to pledge? You can also donate to an effective charity, sign up to our newsletter, read our blog, attend an event, join an effective altruism group, or get in touch if you'd like to discuss anything.

If you have any updated figures or examples to add to this page, please contact us to let us know.

This work is licensed under a Creative Commons Attribution 4.0 International License.